[DISCUSS] Integrate Flink SQL well with Hive ecosystem
Locked
123
123
Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem
 
|
Hi all, I have also shared a design doc on Hive metastore integration that is attached here and also to FLINK-10556[1]. Please kindly review and share your feedback. Thanks, Xuefu
|
Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem
|
|
Hi Xuefu,
Thanks a lot for driving this big effort. I would suggest convert your proposal and design doc into a google doc, and share it on the dev mailing list for the community to review and comment with title like "[DISCUSS] ... Hive integration design ..." . Once approved, we can document it as a FLIP (Flink Improvement Proposals), and use JIRAs to track the implementations. What do you think? Shuyi On Tue, Oct 30, 2018 at 11:32 AM Zhang, Xuefu <[hidden email]> wrote: > Hi all, > > I have also shared a design doc on Hive metastore integration that is > attached here and also to FLINK-10556[1]. Please kindly review and share > your feedback. > > > Thanks, > Xuefu > > [1] https://issues.apache.org/jira/browse/FLINK-10556 > > ------------------------------------------------------------------ > Sender:Xuefu <[hidden email]> > Sent at:2018 Oct 25 (Thu) 01:08 > Recipient:Xuefu <[hidden email]>; Shuyi Chen <[hidden email]> > Cc:yanghua1127 <[hidden email]>; Fabian Hueske <[hidden email]>; > dev <[hidden email]>; user <[hidden email]> > Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem > > Hi all, > > To wrap up the discussion, I have attached a PDF describing the proposal, > which is also attached to FLINK-10556 [1]. Please feel free to watch that > JIRA to track the progress. > > Please also let me know if you have additional comments or questions. > > Thanks, > Xuefu > > [1] https://issues.apache.org/jira/browse/FLINK-10556 > > > ------------------------------------------------------------------ > Sender:Xuefu <[hidden email]> > Sent at:2018 Oct 16 (Tue) 03:40 > Recipient:Shuyi Chen <[hidden email]> > Cc:yanghua1127 <[hidden email]>; Fabian Hueske <[hidden email]>; > dev <[hidden email]>; user <[hidden email]> > Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem > > Hi Shuyi, > > Thank you for your input. Yes, I agreed with a phased approach and like to > move forward fast. :) We did some work internally on DDL utilizing babel > parser in Calcite. While babel makes Calcite's grammar extensible, at > first impression it still seems too cumbersome for a project when too > much extensions are made. It's even challenging to find where the extension > is needed! It would be certainly better if Calcite can magically support > Hive QL by just turning on a flag, such as that for MYSQL_5. I can also > see that this could mean a lot of work on Calcite. Nevertheless, I will > bring up the discussion over there and to see what their community thinks. > > Would mind to share more info about the proposal on DDL that you > mentioned? We can certainly collaborate on this. > > Thanks, > Xuefu > > ------------------------------------------------------------------ > Sender:Shuyi Chen <[hidden email]> > Sent at:2018 Oct 14 (Sun) 08:30 > Recipient:Xuefu <[hidden email]> > Cc:yanghua1127 <[hidden email]>; Fabian Hueske <[hidden email]>; > dev <[hidden email]>; user <[hidden email]> > Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem > > Welcome to the community and thanks for the great proposal, Xuefu! I think > the proposal can be divided into 2 stages: making Flink to support Hive > features, and make Hive to work with Flink. I agreed with Timo that on > starting with a smaller scope, so we can make progress faster. As for [6], > a proposal for DDL is already in progress, and will come after the unified > SQL connector API is done. For supporting Hive syntax, we might need to > work with the Calcite community, and a recent effort called babel ( > https://issues.apache.org/jira/browse/CALCITE-2280) in Calcite might help > here. > > Thanks > Shuyi > > On Wed, Oct 10, 2018 at 8:02 PM Zhang, Xuefu <[hidden email]> > wrote: > Hi Fabian/Vno, > > Thank you very much for your encouragement inquiry. Sorry that I didn't > see Fabian's email until I read Vino's response just now. (Somehow Fabian's > went to the spam folder.) > > My proposal contains long-term and short-terms goals. Nevertheless, the > effort will focus on the following areas, including Fabian's list: > > 1. Hive metastore connectivity - This covers both read/write access, which > means Flink can make full use of Hive's metastore as its catalog (at least > for the batch but can extend for streaming as well). > 2. Metadata compatibility - Objects (databases, tables, partitions, etc) > created by Hive can be understood by Flink and the reverse direction is > true also. > 3. Data compatibility - Similar to #2, data produced by Hive can be > consumed by Flink and vise versa. > 4. Support Hive UDFs - For all Hive's native udfs, Flink either provides > its own implementation or make Hive's implementation work in Flink. > Further, for user created UDFs in Hive, Flink SQL should provide a > mechanism allowing user to import them into Flink without any code change > required. > 5. Data types - Flink SQL should support all data types that are > available in Hive. > 6. SQL Language - Flink SQL should support SQL standard (such as SQL2003) > with extension to support Hive's syntax and language features, around DDL, > DML, and SELECT queries. > 7. SQL CLI - this is currently developing in Flink but more effort is > needed. > 8. Server - provide a server that's compatible with Hive's HiverServer2 in > thrift APIs, such that HiveServer2 users can reuse their existing client > (such as beeline) but connect to Flink's thrift server instead. > 9. JDBC/ODBC drivers - Flink may provide its own JDBC/ODBC drivers for > other application to use to connect to its thrift server > 10. Support other user's customizations in Hive, such as Hive Serdes, > storage handlers, etc. > 11. Better task failure tolerance and task scheduling at Flink runtime. > > As you can see, achieving all those requires significant effort and across > all layers in Flink. However, a short-term goal could include only core > areas (such as 1, 2, 4, 5, 6, 7) or start at a smaller scope (such as #3, > #6). > > Please share your further thoughts. If we generally agree that this is the > right direction, I could come up with a formal proposal quickly and then we > can follow up with broader discussions. > > Thanks, > Xuefu > > > > ------------------------------------------------------------------ > Sender:vino yang <[hidden email]> > Sent at:2018 Oct 11 (Thu) 09:45 > Recipient:Fabian Hueske <[hidden email]> > Cc:dev <[hidden email]>; Xuefu <[hidden email]>; user < > [hidden email]> > Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem > > Hi Xuefu, > > Appreciate this proposal, and like Fabian, it would look better if you can > give more details of the plan. > > Thanks, vino. > > Fabian Hueske <[hidden email]> 于2018年10月10日周三 下午5:27写道： > Hi Xuefu, > > Welcome to the Flink community and thanks for starting this discussion! > Better Hive integration would be really great! > Can you go into details of what you are proposing? I can think of a couple > ways to improve Flink in that regard: > > * Support for Hive UDFs > * Support for Hive metadata catalog > * Support for HiveQL syntax > * ??? > > Best, Fabian > > Am Di., 9. Okt. 2018 um 19:22 Uhr schrieb Zhang, Xuefu < > [hidden email]>: > Hi all, > > Along with the community's effort, inside Alibaba we have explored Flink's > potential as an execution engine not just for stream processing but also > for batch processing. We are encouraged by our findings and have initiated > our effort to make Flink's SQL capabilities full-fledged. When comparing > what's available in Flink to the offerings from competitive data processing > engines, we identified a major gap in Flink: a well integration with Hive > ecosystem. This is crucial to the success of Flink SQL and batch due to the > well-established data ecosystem around Hive. Therefore, we have done some > initial work along this direction but there are still a lot of effort > needed. > > We have two strategies in mind. The first one is to make Flink SQL > full-fledged and well-integrated with Hive ecosystem. This is a similar > approach to what Spark SQL adopted. The second strategy is to make Hive > itself work with Flink, similar to the proposal in [1]. Each approach bears > its pros and cons, but they don’t need to be mutually exclusive with each > targeting at different users and use cases. We believe that both will > promote a much greater adoption of Flink beyond stream processing. > > We have been focused on the first approach and would like to showcase > Flink's batch and SQL capabilities with Flink SQL. However, we have also > planned to start strategy #2 as the follow-up effort. > > I'm completely new to Flink(, with a short bio [2] below), though many of > my colleagues here at Alibaba are long-time contributors. Nevertheless, I'd > like to share our thoughts and invite your early feedback. At the same > time, I am working on a detailed proposal on Flink SQL's integration with > Hive ecosystem, which will be also shared when ready. > > While the ideas are simple, each approach will demand significant effort, > more than what we can afford. Thus, the input and contributions from the > communities are greatly welcome and appreciated. > > Regards, > > > Xuefu > > References: > > [1] https://issues.apache.org/jira/browse/HIVE-10712 > [2] Xuefu Zhang is a long-time open source veteran, worked or working on > many projects under Apache Foundation, of which he is also an honored > member. About 10 years ago he worked in the Hadoop team at Yahoo where the > projects just got started. Later he worked at Cloudera, initiating and > leading the development of Hive on Spark project in the communities and > across many organizations. Prior to joining Alibaba, he worked at Uber > where he promoted Hive on Spark to all Uber's SQL on Hadoop workload and > significantly improved Uber's cluster efficiency. > > > > > -- > "So you have to trust that the dots will somehow connect in your future." > > -- "So you have to trust that the dots will somehow connect in your future." |
Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem
|
|
Hi Shuiyi,
Good idea. Actually the PDF was converted from a google doc. Here is its link: https://docs.google.com/document/d/1SkppRD_rE3uOKSN-LuZCqn4f7dz0zW5aa6T_hBZq5_o/edit?usp=sharing Once we reach an agreement, I can convert it to a FLIP. Thanks, Xuefu ------------------------------------------------------------------ Sender:Shuyi Chen <[hidden email]> Sent at:2018 Nov 1 (Thu) 02:47 Recipient:Xuefu <[hidden email]> Cc:vino yang <[hidden email]>; Fabian Hueske <[hidden email]>; dev <[hidden email]>; user <[hidden email]> Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem Hi Xuefu, Thanks a lot for driving this big effort. I would suggest convert your proposal and design doc into a google doc, and share it on the dev mailing list for the community to review and comment with title like "[DISCUSS] ... Hive integration design ..." . Once approved, we can document it as a FLIP (Flink Improvement Proposals), and use JIRAs to track the implementations. What do you think? Shuyi On Tue, Oct 30, 2018 at 11:32 AM Zhang, Xuefu <[hidden email]> wrote: Hi all, I have also shared a design doc on Hive metastore integration that is attached here and also to FLINK-10556[1]. Please kindly review and share your feedback. Thanks, Xuefu [1] https://issues.apache.org/jira/browse/FLINK-10556 ------------------------------------------------------------------ Sender:Xuefu <[hidden email]> Sent at:2018 Oct 25 (Thu) 01:08 Recipient:Xuefu <[hidden email]>; Shuyi Chen <[hidden email]> Cc:yanghua1127 <[hidden email]>; Fabian Hueske <[hidden email]>; dev <[hidden email]>; user <[hidden email]> Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem Hi all, To wrap up the discussion, I have attached a PDF describing the proposal, which is also attached to FLINK-10556 [1]. Please feel free to watch that JIRA to track the progress. Please also let me know if you have additional comments or questions. Thanks, Xuefu [1] https://issues.apache.org/jira/browse/FLINK-10556 ------------------------------------------------------------------ Sender:Xuefu <[hidden email]> Sent at:2018 Oct 16 (Tue) 03:40 Recipient:Shuyi Chen <[hidden email]> Cc:yanghua1127 <[hidden email]>; Fabian Hueske <[hidden email]>; dev <[hidden email]>; user <[hidden email]> Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem Hi Shuyi, Thank you for your input. Yes, I agreed with a phased approach and like to move forward fast. :) We did some work internally on DDL utilizing babel parser in Calcite. While babel makes Calcite's grammar extensible, at first impression it still seems too cumbersome for a project when too much extensions are made. It's even challenging to find where the extension is needed! It would be certainly better if Calcite can magically support Hive QL by just turning on a flag, such as that for MYSQL_5. I can also see that this could mean a lot of work on Calcite. Nevertheless, I will bring up the discussion over there and to see what their community thinks. Would mind to share more info about the proposal on DDL that you mentioned? We can certainly collaborate on this. Thanks, Xuefu ------------------------------------------------------------------ Sender:Shuyi Chen <[hidden email]> Sent at:2018 Oct 14 (Sun) 08:30 Recipient:Xuefu <[hidden email]> Cc:yanghua1127 <[hidden email]>; Fabian Hueske <[hidden email]>; dev <[hidden email]>; user <[hidden email]> Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem Welcome to the community and thanks for the great proposal, Xuefu! I think the proposal can be divided into 2 stages: making Flink to support Hive features, and make Hive to work with Flink. I agreed with Timo that on starting with a smaller scope, so we can make progress faster. As for [6], a proposal for DDL is already in progress, and will come after the unified SQL connector API is done. For supporting Hive syntax, we might need to work with the Calcite community, and a recent effort called babel (https://issues.apache.org/jira/browse/CALCITE-2280) in Calcite might help here. Thanks Shuyi On Wed, Oct 10, 2018 at 8:02 PM Zhang, Xuefu <[hidden email]> wrote: Hi Fabian/Vno, Thank you very much for your encouragement inquiry. Sorry that I didn't see Fabian's email until I read Vino's response just now. (Somehow Fabian's went to the spam folder.) My proposal contains long-term and short-terms goals. Nevertheless, the effort will focus on the following areas, including Fabian's list: 1. Hive metastore connectivity - This covers both read/write access, which means Flink can make full use of Hive's metastore as its catalog (at least for the batch but can extend for streaming as well). 2. Metadata compatibility - Objects (databases, tables, partitions, etc) created by Hive can be understood by Flink and the reverse direction is true also. 3. Data compatibility - Similar to #2, data produced by Hive can be consumed by Flink and vise versa. 4. Support Hive UDFs - For all Hive's native udfs, Flink either provides its own implementation or make Hive's implementation work in Flink. Further, for user created UDFs in Hive, Flink SQL should provide a mechanism allowing user to import them into Flink without any code change required. 5. Data types - Flink SQL should support all data types that are available in Hive. 6. SQL Language - Flink SQL should support SQL standard (such as SQL2003) with extension to support Hive's syntax and language features, around DDL, DML, and SELECT queries. 7. SQL CLI - this is currently developing in Flink but more effort is needed. 8. Server - provide a server that's compatible with Hive's HiverServer2 in thrift APIs, such that HiveServer2 users can reuse their existing client (such as beeline) but connect to Flink's thrift server instead. 9. JDBC/ODBC drivers - Flink may provide its own JDBC/ODBC drivers for other application to use to connect to its thrift server 10. Support other user's customizations in Hive, such as Hive Serdes, storage handlers, etc. 11. Better task failure tolerance and task scheduling at Flink runtime. As you can see, achieving all those requires significant effort and across all layers in Flink. However, a short-term goal could include only core areas (such as 1, 2, 4, 5, 6, 7) or start at a smaller scope (such as #3, #6). Please share your further thoughts. If we generally agree that this is the right direction, I could come up with a formal proposal quickly and then we can follow up with broader discussions. Thanks, Xuefu ------------------------------------------------------------------ Sender:vino yang <[hidden email]> Sent at:2018 Oct 11 (Thu) 09:45 Recipient:Fabian Hueske <[hidden email]> Cc:dev <[hidden email]>; Xuefu <[hidden email]>; user <[hidden email]> Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem Hi Xuefu, Appreciate this proposal, and like Fabian, it would look better if you can give more details of the plan. Thanks, vino. Fabian Hueske <[hidden email]> 于2018年10月10日周三 下午5:27写道： Hi Xuefu, Welcome to the Flink community and thanks for starting this discussion! Better Hive integration would be really great! Can you go into details of what you are proposing? I can think of a couple ways to improve Flink in that regard: * Support for Hive UDFs * Support for Hive metadata catalog * Support for HiveQL syntax * ??? Best, Fabian Am Di., 9. Okt. 2018 um 19:22 Uhr schrieb Zhang, Xuefu <[hidden email]>: Hi all, Along with the community's effort, inside Alibaba we have explored Flink's potential as an execution engine not just for stream processing but also for batch processing. We are encouraged by our findings and have initiated our effort to make Flink's SQL capabilities full-fledged. When comparing what's available in Flink to the offerings from competitive data processing engines, we identified a major gap in Flink: a well integration with Hive ecosystem. This is crucial to the success of Flink SQL and batch due to the well-established data ecosystem around Hive. Therefore, we have done some initial work along this direction but there are still a lot of effort needed. We have two strategies in mind. The first one is to make Flink SQL full-fledged and well-integrated with Hive ecosystem. This is a similar approach to what Spark SQL adopted. The second strategy is to make Hive itself work with Flink, similar to the proposal in [1]. Each approach bears its pros and cons, but they don’t need to be mutually exclusive with each targeting at different users and use cases. We believe that both will promote a much greater adoption of Flink beyond stream processing. We have been focused on the first approach and would like to showcase Flink's batch and SQL capabilities with Flink SQL. However, we have also planned to start strategy #2 as the follow-up effort. I'm completely new to Flink(, with a short bio [2] below), though many of my colleagues here at Alibaba are long-time contributors. Nevertheless, I'd like to share our thoughts and invite your early feedback. At the same time, I am working on a detailed proposal on Flink SQL's integration with Hive ecosystem, which will be also shared when ready. While the ideas are simple, each approach will demand significant effort, more than what we can afford. Thus, the input and contributions from the communities are greatly welcome and appreciated. Regards, Xuefu References: [1] https://issues.apache.org/jira/browse/HIVE-10712 [2] Xuefu Zhang is a long-time open source veteran, worked or working on many projects under Apache Foundation, of which he is also an honored member. About 10 years ago he worked in the Hadoop team at Yahoo where the projects just got started. Later he worked at Cloudera, initiating and leading the development of Hive on Spark project in the communities and across many organizations. Prior to joining Alibaba, he worked at Uber where he promoted Hive on Spark to all Uber's SQL on Hadoop workload and significantly improved Uber's cluster efficiency. -- "So you have to trust that the dots will somehow connect in your future." -- "So you have to trust that the dots will somehow connect in your future." |
|
|
Hi all,
As Xuefu has published the design doc on google, I agree with Shuyi's suggestion that we probably should start a new email thread like "[DISCUSS] ... Hive integration design ..." on only dev mailing list for community devs to review. The current thread sends to both dev and user list. This email thread is more like validating the general idea and direction with the community, and it's been pretty long and crowded so far. Since everyone is pro for the idea, we can move forward with another thread to discuss and finalize the design. Thanks, Bowen On Wed, Oct 31, 2018 at 12:16 PM Zhang, Xuefu <[hidden email]> wrote: > Hi Shuiyi, > > Good idea. Actually the PDF was converted from a google doc. Here is its > link: > > https://docs.google.com/document/d/1SkppRD_rE3uOKSN-LuZCqn4f7dz0zW5aa6T_hBZq5_o/edit?usp=sharing > Once we reach an agreement, I can convert it to a FLIP. > > Thanks, > Xuefu > > > > ------------------------------------------------------------------ > Sender:Shuyi Chen <[hidden email]> > Sent at:2018 Nov 1 (Thu) 02:47 > Recipient:Xuefu <[hidden email]> > Cc:vino yang <[hidden email]>; Fabian Hueske <[hidden email]>; > dev <[hidden email]>; user <[hidden email]> > Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem > > Hi Xuefu, > > Thanks a lot for driving this big effort. I would suggest convert your > proposal and design doc into a google doc, and share it on the dev mailing > list for the community to review and comment with title like "[DISCUSS] ... > Hive integration design ..." . Once approved, we can document it as a FLIP > (Flink Improvement Proposals), and use JIRAs to track the implementations. > What do you think? > > Shuyi > > On Tue, Oct 30, 2018 at 11:32 AM Zhang, Xuefu <[hidden email]> > wrote: > Hi all, > > I have also shared a design doc on Hive metastore integration that is > attached here and also to FLINK-10556[1]. Please kindly review and share > your feedback. > > > Thanks, > Xuefu > > [1] https://issues.apache.org/jira/browse/FLINK-10556 > ------------------------------------------------------------------ > Sender:Xuefu <[hidden email]> > Sent at:2018 Oct 25 (Thu) 01:08 > Recipient:Xuefu <[hidden email]>; Shuyi Chen <[hidden email]> > Cc:yanghua1127 <[hidden email]>; Fabian Hueske <[hidden email]>; > dev <[hidden email]>; user <[hidden email]> > Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem > > Hi all, > > To wrap up the discussion, I have attached a PDF describing the proposal, > which is also attached to FLINK-10556 [1]. Please feel free to watch that > JIRA to track the progress. > > Please also let me know if you have additional comments or questions. > > Thanks, > Xuefu > > [1] https://issues.apache.org/jira/browse/FLINK-10556 > > > ------------------------------------------------------------------ > Sender:Xuefu <[hidden email]> > Sent at:2018 Oct 16 (Tue) 03:40 > Recipient:Shuyi Chen <[hidden email]> > Cc:yanghua1127 <[hidden email]>; Fabian Hueske <[hidden email]>; > dev <[hidden email]>; user <[hidden email]> > Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem > > Hi Shuyi, > > Thank you for your input. Yes, I agreed with a phased approach and like to > move forward fast. :) We did some work internally on DDL utilizing babel > parser in Calcite. While babel makes Calcite's grammar extensible, at > first impression it still seems too cumbersome for a project when too > much extensions are made. It's even challenging to find where the extension > is needed! It would be certainly better if Calcite can magically support > Hive QL by just turning on a flag, such as that for MYSQL_5. I can also > see that this could mean a lot of work on Calcite. Nevertheless, I will > bring up the discussion over there and to see what their community thinks. > > Would mind to share more info about the proposal on DDL that you > mentioned? We can certainly collaborate on this. > > Thanks, > Xuefu > > ------------------------------------------------------------------ > Sender:Shuyi Chen <[hidden email]> > Sent at:2018 Oct 14 (Sun) 08:30 > Recipient:Xuefu <[hidden email]> > Cc:yanghua1127 <[hidden email]>; Fabian Hueske <[hidden email]>; > dev <[hidden email]>; user <[hidden email]> > Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem > > Welcome to the community and thanks for the great proposal, Xuefu! I think > the proposal can be divided into 2 stages: making Flink to support Hive > features, and make Hive to work with Flink. I agreed with Timo that on > starting with a smaller scope, so we can make progress faster. As for [6], > a proposal for DDL is already in progress, and will come after the unified > SQL connector API is done. For supporting Hive syntax, we might need to > work with the Calcite community, and a recent effort called babel ( > https://issues.apache.org/jira/browse/CALCITE-2280) in Calcite might help > here. > > Thanks > Shuyi > > On Wed, Oct 10, 2018 at 8:02 PM Zhang, Xuefu <[hidden email]> > wrote: > Hi Fabian/Vno, > > Thank you very much for your encouragement inquiry. Sorry that I didn't > see Fabian's email until I read Vino's response just now. (Somehow Fabian's > went to the spam folder.) > > My proposal contains long-term and short-terms goals. Nevertheless, the > effort will focus on the following areas, including Fabian's list: > > 1. Hive metastore connectivity - This covers both read/write access, which > means Flink can make full use of Hive's metastore as its catalog (at least > for the batch but can extend for streaming as well). > 2. Metadata compatibility - Objects (databases, tables, partitions, etc) > created by Hive can be understood by Flink and the reverse direction is > true also. > 3. Data compatibility - Similar to #2, data produced by Hive can be > consumed by Flink and vise versa. > 4. Support Hive UDFs - For all Hive's native udfs, Flink either provides > its own implementation or make Hive's implementation work in Flink. > Further, for user created UDFs in Hive, Flink SQL should provide a > mechanism allowing user to import them into Flink without any code change > required. > 5. Data types - Flink SQL should support all data types that are > available in Hive. > 6. SQL Language - Flink SQL should support SQL standard (such as SQL2003) > with extension to support Hive's syntax and language features, around DDL, > DML, and SELECT queries. > 7. SQL CLI - this is currently developing in Flink but more effort is > needed. > 8. Server - provide a server that's compatible with Hive's HiverServer2 in > thrift APIs, such that HiveServer2 users can reuse their existing client > (such as beeline) but connect to Flink's thrift server instead. > 9. JDBC/ODBC drivers - Flink may provide its own JDBC/ODBC drivers for > other application to use to connect to its thrift server > 10. Support other user's customizations in Hive, such as Hive Serdes, > storage handlers, etc. > 11. Better task failure tolerance and task scheduling at Flink runtime. > > As you can see, achieving all those requires significant effort and across > all layers in Flink. However, a short-term goal could include only core > areas (such as 1, 2, 4, 5, 6, 7) or start at a smaller scope (such as #3, > #6). > > Please share your further thoughts. If we generally agree that this is the > right direction, I could come up with a formal proposal quickly and then we > can follow up with broader discussions. > > Thanks, > Xuefu > > > > ------------------------------------------------------------------ > Sender:vino yang <[hidden email]> > Sent at:2018 Oct 11 (Thu) 09:45 > Recipient:Fabian Hueske <[hidden email]> > Cc:dev <[hidden email]>; Xuefu <[hidden email]>; user < > [hidden email]> > Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem > > Hi Xuefu, > > Appreciate this proposal, and like Fabian, it would look better if you can > give more details of the plan. > > Thanks, vino. > > Fabian Hueske <[hidden email]> 于2018年10月10日周三 下午5:27写道： > Hi Xuefu, > > Welcome to the Flink community and thanks for starting this discussion! > Better Hive integration would be really great! > Can you go into details of what you are proposing? I can think of a couple > ways to improve Flink in that regard: > > * Support for Hive UDFs > * Support for Hive metadata catalog > * Support for HiveQL syntax > * ??? > > Best, Fabian > > Am Di., 9. Okt. 2018 um 19:22 Uhr schrieb Zhang, Xuefu < > [hidden email]>: > Hi all, > > Along with the community's effort, inside Alibaba we have explored Flink's > potential as an execution engine not just for stream processing but also > for batch processing. We are encouraged by our findings and have initiated > our effort to make Flink's SQL capabilities full-fledged. When comparing > what's available in Flink to the offerings from competitive data processing > engines, we identified a major gap in Flink: a well integration with Hive > ecosystem. This is crucial to the success of Flink SQL and batch due to the > well-established data ecosystem around Hive. Therefore, we have done some > initial work along this direction but there are still a lot of effort > needed. > > We have two strategies in mind. The first one is to make Flink SQL > full-fledged and well-integrated with Hive ecosystem. This is a similar > approach to what Spark SQL adopted. The second strategy is to make Hive > itself work with Flink, similar to the proposal in [1]. Each approach bears > its pros and cons, but they don’t need to be mutually exclusive with each > targeting at different users and use cases. We believe that both will > promote a much greater adoption of Flink beyond stream processing. > > We have been focused on the first approach and would like to showcase > Flink's batch and SQL capabilities with Flink SQL. However, we have also > planned to start strategy #2 as the follow-up effort. > > I'm completely new to Flink(, with a short bio [2] below), though many of > my colleagues here at Alibaba are long-time contributors. Nevertheless, I'd > like to share our thoughts and invite your early feedback. At the same > time, I am working on a detailed proposal on Flink SQL's integration with > Hive ecosystem, which will be also shared when ready. > > While the ideas are simple, each approach will demand significant effort, > more than what we can afford. Thus, the input and contributions from the > communities are greatly welcome and appreciated. > > Regards, > > > Xuefu > > References: > > [1] https://issues.apache.org/jira/browse/HIVE-10712 > [2] Xuefu Zhang is a long-time open source veteran, worked or working on > many projects under Apache Foundation, of which he is also an honored > member. About 10 years ago he worked in the Hadoop team at Yahoo where the > projects just got started. Later he worked at Cloudera, initiating and > leading the development of Hive on Spark project in the communities and > across many organizations. Prior to joining Alibaba, he worked at Uber > where he promoted Hive on Spark to all Uber's SQL on Hadoop workload and > significantly improved Uber's cluster efficiency. > > > > > -- > "So you have to trust that the dots will somehow connect in your future." > > > -- > "So you have to trust that the dots will somehow connect in your future." > > |
|
|
After taking a look at how other discussion threads work, I think it's
actually fine just keep our discussion here. It's up to you, Xuefu. The google doc LGTM. I left some minor comments. On Thu, Nov 1, 2018 at 10:17 AM Bowen Li <[hidden email]> wrote: > Hi all, > > As Xuefu has published the design doc on google, I agree with Shuyi's > suggestion that we probably should start a new email thread like "[DISCUSS] > ... Hive integration design ..." on only dev mailing list for community > devs to review. The current thread sends to both dev and user list. > > This email thread is more like validating the general idea and direction > with the community, and it's been pretty long and crowded so far. Since > everyone is pro for the idea, we can move forward with another thread to > discuss and finalize the design. > > Thanks, > Bowen > > On Wed, Oct 31, 2018 at 12:16 PM Zhang, Xuefu <[hidden email]> > wrote: > >> Hi Shuiyi, >> >> Good idea. Actually the PDF was converted from a google doc. Here is its >> link: >> >> https://docs.google.com/document/d/1SkppRD_rE3uOKSN-LuZCqn4f7dz0zW5aa6T_hBZq5_o/edit?usp=sharing >> Once we reach an agreement, I can convert it to a FLIP. >> >> Thanks, >> Xuefu >> >> >> >> ------------------------------------------------------------------ >> Sender:Shuyi Chen <[hidden email]> >> Sent at:2018 Nov 1 (Thu) 02:47 >> Recipient:Xuefu <[hidden email]> >> Cc:vino yang <[hidden email]>; Fabian Hueske <[hidden email]>; >> dev <[hidden email]>; user <[hidden email]> >> Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem >> >> Hi Xuefu, >> >> Thanks a lot for driving this big effort. I would suggest convert your >> proposal and design doc into a google doc, and share it on the dev mailing >> list for the community to review and comment with title like "[DISCUSS] ... >> Hive integration design ..." . Once approved, we can document it as a FLIP >> (Flink Improvement Proposals), and use JIRAs to track the implementations. >> What do you think? >> >> Shuyi >> >> On Tue, Oct 30, 2018 at 11:32 AM Zhang, Xuefu <[hidden email]> >> wrote: >> Hi all, >> >> I have also shared a design doc on Hive metastore integration that is >> attached here and also to FLINK-10556[1]. Please kindly review and share >> your feedback. >> >> >> Thanks, >> Xuefu >> >> [1] https://issues.apache.org/jira/browse/FLINK-10556 >> ------------------------------------------------------------------ >> Sender:Xuefu <[hidden email]> >> Sent at:2018 Oct 25 (Thu) 01:08 >> Recipient:Xuefu <[hidden email]>; Shuyi Chen <[hidden email] >> > >> Cc:yanghua1127 <[hidden email]>; Fabian Hueske <[hidden email]>; >> dev <[hidden email]>; user <[hidden email]> >> Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem >> >> Hi all, >> >> To wrap up the discussion, I have attached a PDF describing the proposal, >> which is also attached to FLINK-10556 [1]. Please feel free to watch that >> JIRA to track the progress. >> >> Please also let me know if you have additional comments or questions. >> >> Thanks, >> Xuefu >> >> [1] https://issues.apache.org/jira/browse/FLINK-10556 >> >> >> ------------------------------------------------------------------ >> Sender:Xuefu <[hidden email]> >> Sent at:2018 Oct 16 (Tue) 03:40 >> Recipient:Shuyi Chen <[hidden email]> >> Cc:yanghua1127 <[hidden email]>; Fabian Hueske <[hidden email]>; >> dev <[hidden email]>; user <[hidden email]> >> Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem >> >> Hi Shuyi, >> >> Thank you for your input. Yes, I agreed with a phased approach and like >> to move forward fast. :) We did some work internally on DDL utilizing babel >> parser in Calcite. While babel makes Calcite's grammar extensible, at >> first impression it still seems too cumbersome for a project when too >> much extensions are made. It's even challenging to find where the extension >> is needed! It would be certainly better if Calcite can magically support >> Hive QL by just turning on a flag, such as that for MYSQL_5. I can also >> see that this could mean a lot of work on Calcite. Nevertheless, I will >> bring up the discussion over there and to see what their community thinks. >> >> Would mind to share more info about the proposal on DDL that you >> mentioned? We can certainly collaborate on this. >> >> Thanks, >> Xuefu >> >> ------------------------------------------------------------------ >> Sender:Shuyi Chen <[hidden email]> >> Sent at:2018 Oct 14 (Sun) 08:30 >> Recipient:Xuefu <[hidden email]> >> Cc:yanghua1127 <[hidden email]>; Fabian Hueske <[hidden email]>; >> dev <[hidden email]>; user <[hidden email]> >> Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem >> >> Welcome to the community and thanks for the great proposal, Xuefu! I >> think the proposal can be divided into 2 stages: making Flink to support >> Hive features, and make Hive to work with Flink. I agreed with Timo that on >> starting with a smaller scope, so we can make progress faster. As for [6], >> a proposal for DDL is already in progress, and will come after the unified >> SQL connector API is done. For supporting Hive syntax, we might need to >> work with the Calcite community, and a recent effort called babel ( >> https://issues.apache.org/jira/browse/CALCITE-2280) in Calcite might >> help here. >> >> Thanks >> Shuyi >> >> On Wed, Oct 10, 2018 at 8:02 PM Zhang, Xuefu <[hidden email]> >> wrote: >> Hi Fabian/Vno, >> >> Thank you very much for your encouragement inquiry. Sorry that I didn't >> see Fabian's email until I read Vino's response just now. (Somehow Fabian's >> went to the spam folder.) >> >> My proposal contains long-term and short-terms goals. Nevertheless, the >> effort will focus on the following areas, including Fabian's list: >> >> 1. Hive metastore connectivity - This covers both read/write access, >> which means Flink can make full use of Hive's metastore as its catalog (at >> least for the batch but can extend for streaming as well). >> 2. Metadata compatibility - Objects (databases, tables, partitions, etc) >> created by Hive can be understood by Flink and the reverse direction is >> true also. >> 3. Data compatibility - Similar to #2, data produced by Hive can be >> consumed by Flink and vise versa. >> 4. Support Hive UDFs - For all Hive's native udfs, Flink either provides >> its own implementation or make Hive's implementation work in Flink. >> Further, for user created UDFs in Hive, Flink SQL should provide a >> mechanism allowing user to import them into Flink without any code change >> required. >> 5. Data types - Flink SQL should support all data types that are >> available in Hive. >> 6. SQL Language - Flink SQL should support SQL standard (such as SQL2003) >> with extension to support Hive's syntax and language features, around DDL, >> DML, and SELECT queries. >> 7. SQL CLI - this is currently developing in Flink but more effort is >> needed. >> 8. Server - provide a server that's compatible with Hive's HiverServer2 >> in thrift APIs, such that HiveServer2 users can reuse their existing client >> (such as beeline) but connect to Flink's thrift server instead. >> 9. JDBC/ODBC drivers - Flink may provide its own JDBC/ODBC drivers for >> other application to use to connect to its thrift server >> 10. Support other user's customizations in Hive, such as Hive Serdes, >> storage handlers, etc. >> 11. Better task failure tolerance and task scheduling at Flink runtime. >> >> As you can see, achieving all those requires significant effort and >> across all layers in Flink. However, a short-term goal could include only >> core areas (such as 1, 2, 4, 5, 6, 7) or start at a smaller scope (such as >> #3, #6). >> >> Please share your further thoughts. If we generally agree that this is >> the right direction, I could come up with a formal proposal quickly and >> then we can follow up with broader discussions. >> >> Thanks, >> Xuefu >> >> >> >> ------------------------------------------------------------------ >> Sender:vino yang <[hidden email]> >> Sent at:2018 Oct 11 (Thu) 09:45 >> Recipient:Fabian Hueske <[hidden email]> >> Cc:dev <[hidden email]>; Xuefu <[hidden email]>; user < >> [hidden email]> >> Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem >> >> Hi Xuefu, >> >> Appreciate this proposal, and like Fabian, it would look better if you >> can give more details of the plan. >> >> Thanks, vino. >> >> Fabian Hueske <[hidden email]> 于2018年10月10日周三 下午5:27写道： >> Hi Xuefu, >> >> Welcome to the Flink community and thanks for starting this discussion! >> Better Hive integration would be really great! >> Can you go into details of what you are proposing? I can think of a >> couple ways to improve Flink in that regard: >> >> * Support for Hive UDFs >> * Support for Hive metadata catalog >> * Support for HiveQL syntax >> * ??? >> >> Best, Fabian >> >> Am Di., 9. Okt. 2018 um 19:22 Uhr schrieb Zhang, Xuefu < >> [hidden email]>: >> Hi all, >> >> Along with the community's effort, inside Alibaba we have explored >> Flink's potential as an execution engine not just for stream processing but >> also for batch processing. We are encouraged by our findings and have >> initiated our effort to make Flink's SQL capabilities full-fledged. When >> comparing what's available in Flink to the offerings from competitive data >> processing engines, we identified a major gap in Flink: a well integration >> with Hive ecosystem. This is crucial to the success of Flink SQL and batch >> due to the well-established data ecosystem around Hive. Therefore, we have >> done some initial work along this direction but there are still a lot of >> effort needed. >> >> We have two strategies in mind. The first one is to make Flink SQL >> full-fledged and well-integrated with Hive ecosystem. This is a similar >> approach to what Spark SQL adopted. The second strategy is to make Hive >> itself work with Flink, similar to the proposal in [1]. Each approach bears >> its pros and cons, but they don’t need to be mutually exclusive with each >> targeting at different users and use cases. We believe that both will >> promote a much greater adoption of Flink beyond stream processing. >> >> We have been focused on the first approach and would like to showcase >> Flink's batch and SQL capabilities with Flink SQL. However, we have also >> planned to start strategy #2 as the follow-up effort. >> >> I'm completely new to Flink(, with a short bio [2] below), though many of >> my colleagues here at Alibaba are long-time contributors. Nevertheless, I'd >> like to share our thoughts and invite your early feedback. At the same >> time, I am working on a detailed proposal on Flink SQL's integration with >> Hive ecosystem, which will be also shared when ready. >> >> While the ideas are simple, each approach will demand significant effort, >> more than what we can afford. Thus, the input and contributions from the >> communities are greatly welcome and appreciated. >> >> Regards, >> >> >> Xuefu >> >> References: >> >> [1] https://issues.apache.org/jira/browse/HIVE-10712 >> [2] Xuefu Zhang is a long-time open source veteran, worked or working on >> many projects under Apache Foundation, of which he is also an honored >> member. About 10 years ago he worked in the Hadoop team at Yahoo where the >> projects just got started. Later he worked at Cloudera, initiating and >> leading the development of Hive on Spark project in the communities and >> across many organizations. Prior to joining Alibaba, he worked at Uber >> where he promoted Hive on Spark to all Uber's SQL on Hadoop workload and >> significantly improved Uber's cluster efficiency. >> >> >> >> >> -- >> "So you have to trust that the dots will somehow connect in your future." >> >> >> -- >> "So you have to trust that the dots will somehow connect in your future." >> >> |
Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem
|
|
Hi Xuefu and all,
Thanks for sharing this design document! I'm very much in favor of restructuring / reworking the catalog handling in Flink SQL as outlined in the document. Most changes described in the design document seem to be rather general and not specifically related to the Hive integration. IMO, there are some aspects, especially those at the boundary of Hive and Flink, that need a bit more discussion. For example * What does it take to make Flink schema compatible with Hive schema? * How will Flink tables (descriptors) be stored in HMS? * How do both Hive catalogs differ? Could they be integrated into to a single one? When to use which one? * What meta information is provided by HMS? What of this can be leveraged by Flink? Thank you, Fabian Am Fr., 2. Nov. 2018 um 00:31 Uhr schrieb Bowen Li <[hidden email]>: > After taking a look at how other discussion threads work, I think it's > actually fine just keep our discussion here. It's up to you, Xuefu. > > The google doc LGTM. I left some minor comments. > > On Thu, Nov 1, 2018 at 10:17 AM Bowen Li <[hidden email]> wrote: > >> Hi all, >> >> As Xuefu has published the design doc on google, I agree with Shuyi's >> suggestion that we probably should start a new email thread like "[DISCUSS] >> ... Hive integration design ..." on only dev mailing list for community >> devs to review. The current thread sends to both dev and user list. >> >> This email thread is more like validating the general idea and direction >> with the community, and it's been pretty long and crowded so far. Since >> everyone is pro for the idea, we can move forward with another thread to >> discuss and finalize the design. >> >> Thanks, >> Bowen >> >> On Wed, Oct 31, 2018 at 12:16 PM Zhang, Xuefu <[hidden email]> >> wrote: >> >>> Hi Shuiyi, >>> >>> Good idea. Actually the PDF was converted from a google doc. Here is its >>> link: >>> >>> https://docs.google.com/document/d/1SkppRD_rE3uOKSN-LuZCqn4f7dz0zW5aa6T_hBZq5_o/edit?usp=sharing >>> Once we reach an agreement, I can convert it to a FLIP. >>> >>> Thanks, >>> Xuefu >>> >>> >>> >>> ------------------------------------------------------------------ >>> Sender:Shuyi Chen <[hidden email]> >>> Sent at:2018 Nov 1 (Thu) 02:47 >>> Recipient:Xuefu <[hidden email]> >>> Cc:vino yang <[hidden email]>; Fabian Hueske <[hidden email]>; >>> dev <[hidden email]>; user <[hidden email]> >>> Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem >>> >>> Hi Xuefu, >>> >>> Thanks a lot for driving this big effort. I would suggest convert your >>> proposal and design doc into a google doc, and share it on the dev mailing >>> list for the community to review and comment with title like "[DISCUSS] ... >>> Hive integration design ..." . Once approved, we can document it as a FLIP >>> (Flink Improvement Proposals), and use JIRAs to track the implementations. >>> What do you think? >>> >>> Shuyi >>> >>> On Tue, Oct 30, 2018 at 11:32 AM Zhang, Xuefu <[hidden email]> >>> wrote: >>> Hi all, >>> >>> I have also shared a design doc on Hive metastore integration that is >>> attached here and also to FLINK-10556[1]. Please kindly review and share >>> your feedback. >>> >>> >>> Thanks, >>> Xuefu >>> >>> [1] https://issues.apache.org/jira/browse/FLINK-10556 >>> ------------------------------------------------------------------ >>> Sender:Xuefu <[hidden email]> >>> Sent at:2018 Oct 25 (Thu) 01:08 >>> Recipient:Xuefu <[hidden email]>; Shuyi Chen < >>> [hidden email]> >>> Cc:yanghua1127 <[hidden email]>; Fabian Hueske <[hidden email]>; >>> dev <[hidden email]>; user <[hidden email]> >>> Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem >>> >>> Hi all, >>> >>> To wrap up the discussion, I have attached a PDF describing the >>> proposal, which is also attached to FLINK-10556 [1]. Please feel free to >>> watch that JIRA to track the progress. >>> >>> Please also let me know if you have additional comments or questions. >>> >>> Thanks, >>> Xuefu >>> >>> [1] https://issues.apache.org/jira/browse/FLINK-10556 >>> >>> >>> ------------------------------------------------------------------ >>> Sender:Xuefu <[hidden email]> >>> Sent at:2018 Oct 16 (Tue) 03:40 >>> Recipient:Shuyi Chen <[hidden email]> >>> Cc:yanghua1127 <[hidden email]>; Fabian Hueske <[hidden email]>; >>> dev <[hidden email]>; user <[hidden email]> >>> Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem >>> >>> Hi Shuyi, >>> >>> Thank you for your input. Yes, I agreed with a phased approach and like >>> to move forward fast. :) We did some work internally on DDL utilizing babel >>> parser in Calcite. While babel makes Calcite's grammar extensible, at >>> first impression it still seems too cumbersome for a project when too >>> much extensions are made. It's even challenging to find where the extension >>> is needed! It would be certainly better if Calcite can magically support >>> Hive QL by just turning on a flag, such as that for MYSQL_5. I can also >>> see that this could mean a lot of work on Calcite. Nevertheless, I will >>> bring up the discussion over there and to see what their community thinks. >>> >>> Would mind to share more info about the proposal on DDL that you >>> mentioned? We can certainly collaborate on this. >>> >>> Thanks, >>> Xuefu >>> >>> ------------------------------------------------------------------ >>> Sender:Shuyi Chen <[hidden email]> >>> Sent at:2018 Oct 14 (Sun) 08:30 >>> Recipient:Xuefu <[hidden email]> >>> Cc:yanghua1127 <[hidden email]>; Fabian Hueske <[hidden email]>; >>> dev <[hidden email]>; user <[hidden email]> >>> Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem >>> >>> Welcome to the community and thanks for the great proposal, Xuefu! I >>> think the proposal can be divided into 2 stages: making Flink to support >>> Hive features, and make Hive to work with Flink. I agreed with Timo that on >>> starting with a smaller scope, so we can make progress faster. As for [6], >>> a proposal for DDL is already in progress, and will come after the unified >>> SQL connector API is done. For supporting Hive syntax, we might need to >>> work with the Calcite community, and a recent effort called babel ( >>> https://issues.apache.org/jira/browse/CALCITE-2280) in Calcite might >>> help here. >>> >>> Thanks >>> Shuyi >>> >>> On Wed, Oct 10, 2018 at 8:02 PM Zhang, Xuefu <[hidden email]> >>> wrote: >>> Hi Fabian/Vno, >>> >>> Thank you very much for your encouragement inquiry. Sorry that I didn't >>> see Fabian's email until I read Vino's response just now. (Somehow Fabian's >>> went to the spam folder.) >>> >>> My proposal contains long-term and short-terms goals. Nevertheless, the >>> effort will focus on the following areas, including Fabian's list: >>> >>> 1. Hive metastore connectivity - This covers both read/write access, >>> which means Flink can make full use of Hive's metastore as its catalog (at >>> least for the batch but can extend for streaming as well). >>> 2. Metadata compatibility - Objects (databases, tables, partitions, etc) >>> created by Hive can be understood by Flink and the reverse direction is >>> true also. >>> 3. Data compatibility - Similar to #2, data produced by Hive can be >>> consumed by Flink and vise versa. >>> 4. Support Hive UDFs - For all Hive's native udfs, Flink either provides >>> its own implementation or make Hive's implementation work in Flink. >>> Further, for user created UDFs in Hive, Flink SQL should provide a >>> mechanism allowing user to import them into Flink without any code change >>> required. >>> 5. Data types - Flink SQL should support all data types that are >>> available in Hive. >>> 6. SQL Language - Flink SQL should support SQL standard (such as >>> SQL2003) with extension to support Hive's syntax and language features, >>> around DDL, DML, and SELECT queries. >>> 7. SQL CLI - this is currently developing in Flink but more effort is >>> needed. >>> 8. Server - provide a server that's compatible with Hive's HiverServer2 >>> in thrift APIs, such that HiveServer2 users can reuse their existing client >>> (such as beeline) but connect to Flink's thrift server instead. >>> 9. JDBC/ODBC drivers - Flink may provide its own JDBC/ODBC drivers for >>> other application to use to connect to its thrift server >>> 10. Support other user's customizations in Hive, such as Hive Serdes, >>> storage handlers, etc. >>> 11. Better task failure tolerance and task scheduling at Flink runtime. >>> >>> As you can see, achieving all those requires significant effort and >>> across all layers in Flink. However, a short-term goal could include only >>> core areas (such as 1, 2, 4, 5, 6, 7) or start at a smaller scope (such as >>> #3, #6). >>> >>> Please share your further thoughts. If we generally agree that this is >>> the right direction, I could come up with a formal proposal quickly and >>> then we can follow up with broader discussions. >>> >>> Thanks, >>> Xuefu >>> >>> >>> >>> ------------------------------------------------------------------ >>> Sender:vino yang <[hidden email]> >>> Sent at:2018 Oct 11 (Thu) 09:45 >>> Recipient:Fabian Hueske <[hidden email]> >>> Cc:dev <[hidden email]>; Xuefu <[hidden email]>; user < >>> [hidden email]> >>> Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem >>> >>> Hi Xuefu, >>> >>> Appreciate this proposal, and like Fabian, it would look better if you >>> can give more details of the plan. >>> >>> Thanks, vino. >>> >>> Fabian Hueske <[hidden email]> 于2018年10月10日周三 下午5:27写道： >>> Hi Xuefu, >>> >>> Welcome to the Flink community and thanks for starting this discussion! >>> Better Hive integration would be really great! >>> Can you go into details of what you are proposing? I can think of a >>> couple ways to improve Flink in that regard: >>> >>> * Support for Hive UDFs >>> * Support for Hive metadata catalog >>> * Support for HiveQL syntax >>> * ??? >>> >>> Best, Fabian >>> >>> Am Di., 9. Okt. 2018 um 19:22 Uhr schrieb Zhang, Xuefu < >>> [hidden email]>: >>> Hi all, >>> >>> Along with the community's effort, inside Alibaba we have explored >>> Flink's potential as an execution engine not just for stream processing but >>> also for batch processing. We are encouraged by our findings and have >>> initiated our effort to make Flink's SQL capabilities full-fledged. When >>> comparing what's available in Flink to the offerings from competitive data >>> processing engines, we identified a major gap in Flink: a well integration >>> with Hive ecosystem. This is crucial to the success of Flink SQL and batch >>> due to the well-established data ecosystem around Hive. Therefore, we have >>> done some initial work along this direction but there are still a lot of >>> effort needed. >>> >>> We have two strategies in mind. The first one is to make Flink SQL >>> full-fledged and well-integrated with Hive ecosystem. This is a similar >>> approach to what Spark SQL adopted. The second strategy is to make Hive >>> itself work with Flink, similar to the proposal in [1]. Each approach bears >>> its pros and cons, but they don’t need to be mutually exclusive with each >>> targeting at different users and use cases. We believe that both will >>> promote a much greater adoption of Flink beyond stream processing. >>> >>> We have been focused on the first approach and would like to showcase >>> Flink's batch and SQL capabilities with Flink SQL. However, we have also >>> planned to start strategy #2 as the follow-up effort. >>> >>> I'm completely new to Flink(, with a short bio [2] below), though many >>> of my colleagues here at Alibaba are long-time contributors. Nevertheless, >>> I'd like to share our thoughts and invite your early feedback. At the same >>> time, I am working on a detailed proposal on Flink SQL's integration with >>> Hive ecosystem, which will be also shared when ready. >>> >>> While the ideas are simple, each approach will demand significant >>> effort, more than what we can afford. Thus, the input and contributions >>> from the communities are greatly welcome and appreciated. >>> >>> Regards, >>> >>> >>> Xuefu >>> >>> References: >>> >>> [1] https://issues.apache.org/jira/browse/HIVE-10712 >>> [2] Xuefu Zhang is a long-time open source veteran, worked or working on >>> many projects under Apache Foundation, of which he is also an honored >>> member. About 10 years ago he worked in the Hadoop team at Yahoo where the >>> projects just got started. Later he worked at Cloudera, initiating and >>> leading the development of Hive on Spark project in the communities and >>> across many organizations. Prior to joining Alibaba, he worked at Uber >>> where he promoted Hive on Spark to all Uber's SQL on Hadoop workload and >>> significantly improved Uber's cluster efficiency. >>> >>> >>> >>> >>> -- >>> "So you have to trust that the dots will somehow connect in your future." >>> >>> >>> -- >>> "So you have to trust that the dots will somehow connect in your future." >>> >>> |
Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem
|
|
Hi,
Maybe we should split this topic (and the design doc) into couple of smaller ones, hopefully independent. The questions that you have asked Fabian have for example very little to do with reading metadata from Hive Meta Store? Piotrek > On 7 Nov 2018, at 14:27, Fabian Hueske <[hidden email]> wrote: > > Hi Xuefu and all, > > Thanks for sharing this design document! > I'm very much in favor of restructuring / reworking the catalog handling in > Flink SQL as outlined in the document. > Most changes described in the design document seem to be rather general and > not specifically related to the Hive integration. > > IMO, there are some aspects, especially those at the boundary of Hive and > Flink, that need a bit more discussion. For example > > * What does it take to make Flink schema compatible with Hive schema? > * How will Flink tables (descriptors) be stored in HMS? > * How do both Hive catalogs differ? Could they be integrated into to a > single one? When to use which one? > * What meta information is provided by HMS? What of this can be leveraged > by Flink? > > Thank you, > Fabian > > Am Fr., 2. Nov. 2018 um 00:31 Uhr schrieb Bowen Li <[hidden email]>: > >> After taking a look at how other discussion threads work, I think it's >> actually fine just keep our discussion here. It's up to you, Xuefu. >> >> The google doc LGTM. I left some minor comments. >> >> On Thu, Nov 1, 2018 at 10:17 AM Bowen Li <[hidden email]> wrote: >> >>> Hi all, >>> >>> As Xuefu has published the design doc on google, I agree with Shuyi's >>> suggestion that we probably should start a new email thread like "[DISCUSS] >>> ... Hive integration design ..." on only dev mailing list for community >>> devs to review. The current thread sends to both dev and user list. >>> >>> This email thread is more like validating the general idea and direction >>> with the community, and it's been pretty long and crowded so far. Since >>> everyone is pro for the idea, we can move forward with another thread to >>> discuss and finalize the design. >>> >>> Thanks, >>> Bowen >>> >>> On Wed, Oct 31, 2018 at 12:16 PM Zhang, Xuefu <[hidden email]> >>> wrote: >>> >>>> Hi Shuiyi, >>>> >>>> Good idea. Actually the PDF was converted from a google doc. Here is its >>>> link: >>>> >>>> https://docs.google.com/document/d/1SkppRD_rE3uOKSN-LuZCqn4f7dz0zW5aa6T_hBZq5_o/edit?usp=sharing >>>> Once we reach an agreement, I can convert it to a FLIP. >>>> >>>> Thanks, >>>> Xuefu >>>> >>>> >>>> >>>> ------------------------------------------------------------------ >>>> Sender:Shuyi Chen <[hidden email]> >>>> Sent at:2018 Nov 1 (Thu) 02:47 >>>> Recipient:Xuefu <[hidden email]> >>>> Cc:vino yang <[hidden email]>; Fabian Hueske <[hidden email]>; >>>> dev <[hidden email]>; user <[hidden email]> >>>> Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem >>>> >>>> Hi Xuefu, >>>> >>>> Thanks a lot for driving this big effort. I would suggest convert your >>>> proposal and design doc into a google doc, and share it on the dev mailing >>>> list for the community to review and comment with title like "[DISCUSS] ... >>>> Hive integration design ..." . Once approved, we can document it as a FLIP >>>> (Flink Improvement Proposals), and use JIRAs to track the implementations. >>>> What do you think? >>>> >>>> Shuyi >>>> >>>> On Tue, Oct 30, 2018 at 11:32 AM Zhang, Xuefu <[hidden email]> >>>> wrote: >>>> Hi all, >>>> >>>> I have also shared a design doc on Hive metastore integration that is >>>> attached here and also to FLINK-10556[1]. Please kindly review and share >>>> your feedback. >>>> >>>> >>>> Thanks, >>>> Xuefu >>>> >>>> [1] https://issues.apache.org/jira/browse/FLINK-10556 >>>> ------------------------------------------------------------------ >>>> Sender:Xuefu <[hidden email]> >>>> Sent at:2018 Oct 25 (Thu) 01:08 >>>> Recipient:Xuefu <[hidden email]>; Shuyi Chen < >>>> [hidden email]> >>>> Cc:yanghua1127 <[hidden email]>; Fabian Hueske <[hidden email]>; >>>> dev <[hidden email]>; user <[hidden email]> >>>> Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem >>>> >>>> Hi all, >>>> >>>> To wrap up the discussion, I have attached a PDF describing the >>>> proposal, which is also attached to FLINK-10556 [1]. Please feel free to >>>> watch that JIRA to track the progress. >>>> >>>> Please also let me know if you have additional comments or questions. >>>> >>>> Thanks, >>>> Xuefu >>>> >>>> [1] https://issues.apache.org/jira/browse/FLINK-10556 >>>> >>>> >>>> ------------------------------------------------------------------ >>>> Sender:Xuefu <[hidden email]> >>>> Sent at:2018 Oct 16 (Tue) 03:40 >>>> Recipient:Shuyi Chen <[hidden email]> >>>> Cc:yanghua1127 <[hidden email]>; Fabian Hueske <[hidden email]>; >>>> dev <[hidden email]>; user <[hidden email]> >>>> Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem >>>> >>>> Hi Shuyi, >>>> >>>> Thank you for your input. Yes, I agreed with a phased approach and like >>>> to move forward fast. :) We did some work internally on DDL utilizing babel >>>> parser in Calcite. While babel makes Calcite's grammar extensible, at >>>> first impression it still seems too cumbersome for a project when too >>>> much extensions are made. It's even challenging to find where the extension >>>> is needed! It would be certainly better if Calcite can magically support >>>> Hive QL by just turning on a flag, such as that for MYSQL_5. I can also >>>> see that this could mean a lot of work on Calcite. Nevertheless, I will >>>> bring up the discussion over there and to see what their community thinks. >>>> >>>> Would mind to share more info about the proposal on DDL that you >>>> mentioned? We can certainly collaborate on this. >>>> >>>> Thanks, >>>> Xuefu >>>> >>>> ------------------------------------------------------------------ >>>> Sender:Shuyi Chen <[hidden email]> >>>> Sent at:2018 Oct 14 (Sun) 08:30 >>>> Recipient:Xuefu <[hidden email]> >>>> Cc:yanghua1127 <[hidden email]>; Fabian Hueske <[hidden email]>; >>>> dev <[hidden email]>; user <[hidden email]> >>>> Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem >>>> >>>> Welcome to the community and thanks for the great proposal, Xuefu! I >>>> think the proposal can be divided into 2 stages: making Flink to support >>>> Hive features, and make Hive to work with Flink. I agreed with Timo that on >>>> starting with a smaller scope, so we can make progress faster. As for [6], >>>> a proposal for DDL is already in progress, and will come after the unified >>>> SQL connector API is done. For supporting Hive syntax, we might need to >>>> work with the Calcite community, and a recent effort called babel ( >>>> https://issues.apache.org/jira/browse/CALCITE-2280) in Calcite might >>>> help here. >>>> >>>> Thanks >>>> Shuyi >>>> >>>> On Wed, Oct 10, 2018 at 8:02 PM Zhang, Xuefu <[hidden email]> >>>> wrote: >>>> Hi Fabian/Vno, >>>> >>>> Thank you very much for your encouragement inquiry. Sorry that I didn't >>>> see Fabian's email until I read Vino's response just now. (Somehow Fabian's >>>> went to the spam folder.) >>>> >>>> My proposal contains long-term and short-terms goals. Nevertheless, the >>>> effort will focus on the following areas, including Fabian's list: >>>> >>>> 1. Hive metastore connectivity - This covers both read/write access, >>>> which means Flink can make full use of Hive's metastore as its catalog (at >>>> least for the batch but can extend for streaming as well). >>>> 2. Metadata compatibility - Objects (databases, tables, partitions, etc) >>>> created by Hive can be understood by Flink and the reverse direction is >>>> true also. >>>> 3. Data compatibility - Similar to #2, data produced by Hive can be >>>> consumed by Flink and vise versa. >>>> 4. Support Hive UDFs - For all Hive's native udfs, Flink either provides >>>> its own implementation or make Hive's implementation work in Flink. >>>> Further, for user created UDFs in Hive, Flink SQL should provide a >>>> mechanism allowing user to import them into Flink without any code change >>>> required. >>>> 5. Data types - Flink SQL should support all data types that are >>>> available in Hive. >>>> 6. SQL Language - Flink SQL should support SQL standard (such as >>>> SQL2003) with extension to support Hive's syntax and language features, >>>> around DDL, DML, and SELECT queries. >>>> 7. SQL CLI - this is currently developing in Flink but more effort is >>>> needed. >>>> 8. Server - provide a server that's compatible with Hive's HiverServer2 >>>> in thrift APIs, such that HiveServer2 users can reuse their existing client >>>> (such as beeline) but connect to Flink's thrift server instead. >>>> 9. JDBC/ODBC drivers - Flink may provide its own JDBC/ODBC drivers for >>>> other application to use to connect to its thrift server >>>> 10. Support other user's customizations in Hive, such as Hive Serdes, >>>> storage handlers, etc. >>>> 11. Better task failure tolerance and task scheduling at Flink runtime. >>>> >>>> As you can see, achieving all those requires significant effort and >>>> across all layers in Flink. However, a short-term goal could include only >>>> core areas (such as 1, 2, 4, 5, 6, 7) or start at a smaller scope (such as >>>> #3, #6). >>>> >>>> Please share your further thoughts. If we generally agree that this is >>>> the right direction, I could come up with a formal proposal quickly and >>>> then we can follow up with broader discussions. >>>> >>>> Thanks, >>>> Xuefu >>>> >>>> >>>> >>>> ------------------------------------------------------------------ >>>> Sender:vino yang <[hidden email]> >>>> Sent at:2018 Oct 11 (Thu) 09:45 >>>> Recipient:Fabian Hueske <[hidden email]> >>>> Cc:dev <[hidden email]>; Xuefu <[hidden email]>; user < >>>> [hidden email]> >>>> Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem >>>> >>>> Hi Xuefu, >>>> >>>> Appreciate this proposal, and like Fabian, it would look better if you >>>> can give more details of the plan. >>>> >>>> Thanks, vino. >>>> >>>> Fabian Hueske <[hidden email]> 于2018年10月10日周三 下午5:27写道： >>>> Hi Xuefu, >>>> >>>> Welcome to the Flink community and thanks for starting this discussion! >>>> Better Hive integration would be really great! >>>> Can you go into details of what you are proposing? I can think of a >>>> couple ways to improve Flink in that regard: >>>> >>>> * Support for Hive UDFs >>>> * Support for Hive metadata catalog >>>> * Support for HiveQL syntax >>>> * ??? >>>> >>>> Best, Fabian >>>> >>>> Am Di., 9. Okt. 2018 um 19:22 Uhr schrieb Zhang, Xuefu < >>>> [hidden email]>: >>>> Hi all, >>>> >>>> Along with the community's effort, inside Alibaba we have explored >>>> Flink's potential as an execution engine not just for stream processing but >>>> also for batch processing. We are encouraged by our findings and have >>>> initiated our effort to make Flink's SQL capabilities full-fledged. When >>>> comparing what's available in Flink to the offerings from competitive data >>>> processing engines, we identified a major gap in Flink: a well integration >>>> with Hive ecosystem. This is crucial to the success of Flink SQL and batch >>>> due to the well-established data ecosystem around Hive. Therefore, we have >>>> done some initial work along this direction but there are still a lot of >>>> effort needed. >>>> >>>> We have two strategies in mind. The first one is to make Flink SQL >>>> full-fledged and well-integrated with Hive ecosystem. This is a similar >>>> approach to what Spark SQL adopted. The second strategy is to make Hive >>>> itself work with Flink, similar to the proposal in [1]. Each approach bears >>>> its pros and cons, but they don’t need to be mutually exclusive with each >>>> targeting at different users and use cases. We believe that both will >>>> promote a much greater adoption of Flink beyond stream processing. >>>> >>>> We have been focused on the first approach and would like to showcase >>>> Flink's batch and SQL capabilities with Flink SQL. However, we have also >>>> planned to start strategy #2 as the follow-up effort. >>>> >>>> I'm completely new to Flink(, with a short bio [2] below), though many >>>> of my colleagues here at Alibaba are long-time contributors. Nevertheless, >>>> I'd like to share our thoughts and invite your early feedback. At the same >>>> time, I am working on a detailed proposal on Flink SQL's integration with >>>> Hive ecosystem, which will be also shared when ready. >>>> >>>> While the ideas are simple, each approach will demand significant >>>> effort, more than what we can afford. Thus, the input and contributions >>>> from the communities are greatly welcome and appreciated. >>>> >>>> Regards, >>>> >>>> >>>> Xuefu >>>> >>>> References: >>>> >>>> [1] https://issues.apache.org/jira/browse/HIVE-10712 >>>> [2] Xuefu Zhang is a long-time open source veteran, worked or working on >>>> many projects under Apache Foundation, of which he is also an honored >>>> member. About 10 years ago he worked in the Hadoop team at Yahoo where the >>>> projects just got started. Later he worked at Cloudera, initiating and >>>> leading the development of Hive on Spark project in the communities and >>>> across many organizations. Prior to joining Alibaba, he worked at Uber >>>> where he promoted Hive on Spark to all Uber's SQL on Hadoop workload and >>>> significantly improved Uber's cluster efficiency. >>>> >>>> >>>> >>>> >>>> -- >>>> "So you have to trust that the dots will somehow connect in your future." >>>> >>>> >>>> -- >>>> "So you have to trust that the dots will somehow connect in your future." >>>> >>>> |
Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem
|
|
Hi Piotr,
That seems to be good idea! Since the google doc for the design is currently under extensive review, I will leave it as it is for now. However, I'll convert it to two different FLIPs when the time comes. How does it sound to you? Thanks, Xuefu ------------------------------------------------------------------ Sender:Piotr Nowojski <[hidden email]> Sent at:2018 Nov 9 (Fri) 02:31 Recipient:dev <[hidden email]> Cc:Bowen Li <[hidden email]>; Xuefu <[hidden email]>; Shuyi Chen <[hidden email]> Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem Hi, Maybe we should split this topic (and the design doc) into couple of smaller ones, hopefully independent. The questions that you have asked Fabian have for example very little to do with reading metadata from Hive Meta Store? Piotrek > On 7 Nov 2018, at 14:27, Fabian Hueske <[hidden email]> wrote: > > Hi Xuefu and all, > > Thanks for sharing this design document! > I'm very much in favor of restructuring / reworking the catalog handling in > Flink SQL as outlined in the document. > Most changes described in the design document seem to be rather general and > not specifically related to the Hive integration. > > IMO, there are some aspects, especially those at the boundary of Hive and > Flink, that need a bit more discussion. For example > > * What does it take to make Flink schema compatible with Hive schema? > * How will Flink tables (descriptors) be stored in HMS? > * How do both Hive catalogs differ? Could they be integrated into to a > single one? When to use which one? > * What meta information is provided by HMS? What of this can be leveraged > by Flink? > > Thank you, > Fabian > > Am Fr., 2. Nov. 2018 um 00:31 Uhr schrieb Bowen Li <[hidden email]>: > >> After taking a look at how other discussion threads work, I think it's >> actually fine just keep our discussion here. It's up to you, Xuefu. >> >> The google doc LGTM. I left some minor comments. >> >> On Thu, Nov 1, 2018 at 10:17 AM Bowen Li <[hidden email]> wrote: >> >>> Hi all, >>> >>> As Xuefu has published the design doc on google, I agree with Shuyi's >>> suggestion that we probably should start a new email thread like "[DISCUSS] >>> ... Hive integration design ..." on only dev mailing list for community >>> devs to review. The current thread sends to both dev and user list. >>> >>> This email thread is more like validating the general idea and direction >>> with the community, and it's been pretty long and crowded so far. Since >>> everyone is pro for the idea, we can move forward with another thread to >>> discuss and finalize the design. >>> >>> Thanks, >>> Bowen >>> >>> On Wed, Oct 31, 2018 at 12:16 PM Zhang, Xuefu <[hidden email]> >>> wrote: >>> >>>> Hi Shuiyi, >>>> >>>> Good idea. Actually the PDF was converted from a google doc. Here is its >>>> link: >>>> >>>> https://docs.google.com/document/d/1SkppRD_rE3uOKSN-LuZCqn4f7dz0zW5aa6T_hBZq5_o/edit?usp=sharing >>>> Once we reach an agreement, I can convert it to a FLIP. >>>> >>>> Thanks, >>>> Xuefu >>>> >>>> >>>> >>>> ------------------------------------------------------------------ >>>> Sender:Shuyi Chen <[hidden email]> >>>> Sent at:2018 Nov 1 (Thu) 02:47 >>>> Recipient:Xuefu <[hidden email]> >>>> Cc:vino yang <[hidden email]>; Fabian Hueske <[hidden email]>; >>>> dev <[hidden email]>; user <[hidden email]> >>>> Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem >>>> >>>> Hi Xuefu, >>>> >>>> Thanks a lot for driving this big effort. I would suggest convert your >>>> proposal and design doc into a google doc, and share it on the dev mailing >>>> list for the community to review and comment with title like "[DISCUSS] ... >>>> Hive integration design ..." . Once approved, we can document it as a FLIP >>>> (Flink Improvement Proposals), and use JIRAs to track the implementations. >>>> What do you think? >>>> >>>> Shuyi >>>> >>>> On Tue, Oct 30, 2018 at 11:32 AM Zhang, Xuefu <[hidden email]> >>>> wrote: >>>> Hi all, >>>> >>>> I have also shared a design doc on Hive metastore integration that is >>>> attached here and also to FLINK-10556[1]. Please kindly review and share >>>> your feedback. >>>> >>>> >>>> Thanks, >>>> Xuefu >>>> >>>> [1] https://issues.apache.org/jira/browse/FLINK-10556 >>>> ------------------------------------------------------------------ >>>> Sender:Xuefu <[hidden email]> >>>> Sent at:2018 Oct 25 (Thu) 01:08 >>>> Recipient:Xuefu <[hidden email]>; Shuyi Chen < >>>> [hidden email]> >>>> Cc:yanghua1127 <[hidden email]>; Fabian Hueske <[hidden email]>; >>>> dev <[hidden email]>; user <[hidden email]> >>>> Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem >>>> >>>> Hi all, >>>> >>>> To wrap up the discussion, I have attached a PDF describing the >>>> proposal, which is also attached to FLINK-10556 [1]. Please feel free to >>>> watch that JIRA to track the progress. >>>> >>>> Please also let me know if you have additional comments or questions. >>>> >>>> Thanks, >>>> Xuefu >>>> >>>> [1] https://issues.apache.org/jira/browse/FLINK-10556 >>>> >>>> >>>> ------------------------------------------------------------------ >>>> Sender:Xuefu <[hidden email]> >>>> Sent at:2018 Oct 16 (Tue) 03:40 >>>> Recipient:Shuyi Chen <[hidden email]> >>>> Cc:yanghua1127 <[hidden email]>; Fabian Hueske <[hidden email]>; >>>> dev <[hidden email]>; user <[hidden email]> >>>> Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem >>>> >>>> Hi Shuyi, >>>> >>>> Thank you for your input. Yes, I agreed with a phased approach and like >>>> to move forward fast. :) We did some work internally on DDL utilizing babel >>>> parser in Calcite. While babel makes Calcite's grammar extensible, at >>>> first impression it still seems too cumbersome for a project when too >>>> much extensions are made. It's even challenging to find where the extension >>>> is needed! It would be certainly better if Calcite can magically support >>>> Hive QL by just turning on a flag, such as that for MYSQL_5. I can also >>>> see that this could mean a lot of work on Calcite. Nevertheless, I will >>>> bring up the discussion over there and to see what their community thinks. >>>> >>>> Would mind to share more info about the proposal on DDL that you >>>> mentioned? We can certainly collaborate on this. >>>> >>>> Thanks, >>>> Xuefu >>>> >>>> ------------------------------------------------------------------ >>>> Sender:Shuyi Chen <[hidden email]> >>>> Sent at:2018 Oct 14 (Sun) 08:30 >>>> Recipient:Xuefu <[hidden email]> >>>> Cc:yanghua1127 <[hidden email]>; Fabian Hueske <[hidden email]>; >>>> dev <[hidden email]>; user <[hidden email]> >>>> Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem >>>> >>>> Welcome to the community and thanks for the great proposal, Xuefu! I >>>> think the proposal can be divided into 2 stages: making Flink to support >>>> Hive features, and make Hive to work with Flink. I agreed with Timo that on >>>> starting with a smaller scope, so we can make progress faster. As for [6], >>>> a proposal for DDL is already in progress, and will come after the unified >>>> SQL connector API is done. For supporting Hive syntax, we might need to >>>> work with the Calcite community, and a recent effort called babel ( >>>> https://issues.apache.org/jira/browse/CALCITE-2280) in Calcite might >>>> help here. >>>> >>>> Thanks >>>> Shuyi >>>> >>>> On Wed, Oct 10, 2018 at 8:02 PM Zhang, Xuefu <[hidden email]> >>>> wrote: >>>> Hi Fabian/Vno, >>>> >>>> Thank you very much for your encouragement inquiry. Sorry that I didn't >>>> see Fabian's email until I read Vino's response just now. (Somehow Fabian's >>>> went to the spam folder.) >>>> >>>> My proposal contains long-term and short-terms goals. Nevertheless, the >>>> effort will focus on the following areas, including Fabian's list: >>>> >>>> 1. Hive metastore connectivity - This covers both read/write access, >>>> which means Flink can make full use of Hive's metastore as its catalog (at >>>> least for the batch but can extend for streaming as well). >>>> 2. Metadata compatibility - Objects (databases, tables, partitions, etc) >>>> created by Hive can be understood by Flink and the reverse direction is >>>> true also. >>>> 3. Data compatibility - Similar to #2, data produced by Hive can be >>>> consumed by Flink and vise versa. >>>> 4. Support Hive UDFs - For all Hive's native udfs, Flink either provides >>>> its own implementation or make Hive's implementation work in Flink. >>>> Further, for user created UDFs in Hive, Flink SQL should provide a >>>> mechanism allowing user to import them into Flink without any code change >>>> required. >>>> 5. Data types - Flink SQL should support all data types that are >>>> available in Hive. >>>> 6. SQL Language - Flink SQL should support SQL standard (such as >>>> SQL2003) with extension to support Hive's syntax and language features, >>>> around DDL, DML, and SELECT queries. >>>> 7. SQL CLI - this is currently developing in Flink but more effort is >>>> needed. >>>> 8. Server - provide a server that's compatible with Hive's HiverServer2 >>>> in thrift APIs, such that HiveServer2 users can reuse their existing client >>>> (such as beeline) but connect to Flink's thrift server instead. >>>> 9. JDBC/ODBC drivers - Flink may provide its own JDBC/ODBC drivers for >>>> other application to use to connect to its thrift server >>>> 10. Support other user's customizations in Hive, such as Hive Serdes, >>>> storage handlers, etc. >>>> 11. Better task failure tolerance and task scheduling at Flink runtime. >>>> >>>> As you can see, achieving all those requires significant effort and >>>> across all layers in Flink. However, a short-term goal could include only >>>> core areas (such as 1, 2, 4, 5, 6, 7) or start at a smaller scope (such as >>>> #3, #6). >>>> >>>> Please share your further thoughts. If we generally agree that this is >>>> the right direction, I could come up with a formal proposal quickly and >>>> then we can follow up with broader discussions. >>>> >>>> Thanks, >>>> Xuefu >>>> >>>> >>>> >>>> ------------------------------------------------------------------ >>>> Sender:vino yang <[hidden email]> >>>> Sent at:2018 Oct 11 (Thu) 09:45 >>>> Recipient:Fabian Hueske <[hidden email]> >>>> Cc:dev <[hidden email]>; Xuefu <[hidden email]>; user < >>>> [hidden email]> >>>> Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem >>>> >>>> Hi Xuefu, >>>> >>>> Appreciate this proposal, and like Fabian, it would look better if you >>>> can give more details of the plan. >>>> >>>> Thanks, vino. >>>> >>>> Fabian Hueske <[hidden email]> 于2018年10月10日周三 下午5:27写道： >>>> Hi Xuefu, >>>> >>>> Welcome to the Flink community and thanks for starting this discussion! >>>> Better Hive integration would be really great! >>>> Can you go into details of what you are proposing? I can think of a >>>> couple ways to improve Flink in that regard: >>>> >>>> * Support for Hive UDFs >>>> * Support for Hive metadata catalog >>>> * Support for HiveQL syntax >>>> * ??? >>>> >>>> Best, Fabian >>>> >>>> Am Di., 9. Okt. 2018 um 19:22 Uhr schrieb Zhang, Xuefu < >>>> [hidden email]>: >>>> Hi all, >>>> >>>> Along with the community's effort, inside Alibaba we have explored >>>> Flink's potential as an execution engine not just for stream processing but >>>> also for batch processing. We are encouraged by our findings and have >>>> initiated our effort to make Flink's SQL capabilities full-fledged. When >>>> comparing what's available in Flink to the offerings from competitive data >>>> processing engines, we identified a major gap in Flink: a well integration >>>> with Hive ecosystem. This is crucial to the success of Flink SQL and batch >>>> due to the well-established data ecosystem around Hive. Therefore, we have >>>> done some initial work along this direction but there are still a lot of >>>> effort needed. >>>> >>>> We have two strategies in mind. The first one is to make Flink SQL >>>> full-fledged and well-integrated with Hive ecosystem. This is a similar >>>> approach to what Spark SQL adopted. The second strategy is to make Hive >>>> itself work with Flink, similar to the proposal in [1]. Each approach bears >>>> its pros and cons, but they don’t need to be mutually exclusive with each >>>> targeting at different users and use cases. We believe that both will >>>> promote a much greater adoption of Flink beyond stream processing. >>>> >>>> We have been focused on the first approach and would like to showcase >>>> Flink's batch and SQL capabilities with Flink SQL. However, we have also >>>> planned to start strategy #2 as the follow-up effort. >>>> >>>> I'm completely new to Flink(, with a short bio [2] below), though many >>>> of my colleagues here at Alibaba are long-time contributors. Nevertheless, >>>> I'd like to share our thoughts and invite your early feedback. At the same >>>> time, I am working on a detailed proposal on Flink SQL's integration with >>>> Hive ecosystem, which will be also shared when ready. >>>> >>>> While the ideas are simple, each approach will demand significant >>>> effort, more than what we can afford. Thus, the input and contributions >>>> from the communities are greatly welcome and appreciated. >>>> >>>> Regards, >>>> >>>> >>>> Xuefu >>>> >>>> References: >>>> >>>> [1] https://issues.apache.org/jira/browse/HIVE-10712 >>>> [2] Xuefu Zhang is a long-time open source veteran, worked or working on >>>> many projects under Apache Foundation, of which he is also an honored >>>> member. About 10 years ago he worked in the Hadoop team at Yahoo where the >>>> projects just got started. Later he worked at Cloudera, initiating and >>>> leading the development of Hive on Spark project in the communities and >>>> across many organizations. Prior to joining Alibaba, he worked at Uber >>>> where he promoted Hive on Spark to all Uber's SQL on Hadoop workload and >>>> significantly improved Uber's cluster efficiency. >>>> >>>> >>>> >>>> >>>> -- >>>> "So you have to trust that the dots will somehow connect in your future." >>>> >>>> >>>> -- >>>> "So you have to trust that the dots will somehow connect in your future." >>>> >>>> |
Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem
|
|
Hi,
Yes, it seems like the best solution. Maybe someone else can also suggests if we can split it further? Maybe changes in the interface in one doc, reading from hive meta store another and final storing our meta informations in hive meta store? Piotrek > On 9 Nov 2018, at 01:44, Zhang, Xuefu <[hidden email]> wrote: > > Hi Piotr, > > That seems to be good idea! > > Since the google doc for the design is currently under extensive review, I will leave it as it is for now. However, I'll convert it to two different FLIPs when the time comes. > > How does it sound to you? > > Thanks, > Xuefu > > > ------------------------------------------------------------------ > Sender:Piotr Nowojski <[hidden email]> > Sent at:2018 Nov 9 (Fri) 02:31 > Recipient:dev <[hidden email]> > Cc:Bowen Li <[hidden email]>; Xuefu <[hidden email]>; Shuyi Chen <[hidden email]> > Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem > > Hi, > > Maybe we should split this topic (and the design doc) into couple of smaller ones, hopefully independent. The questions that you have asked Fabian have for example very little to do with reading metadata from Hive Meta Store? > > Piotrek > >> On 7 Nov 2018, at 14:27, Fabian Hueske <[hidden email]> wrote: >> >> Hi Xuefu and all, >> >> Thanks for sharing this design document! >> I'm very much in favor of restructuring / reworking the catalog handling in >> Flink SQL as outlined in the document. >> Most changes described in the design document seem to be rather general and >> not specifically related to the Hive integration. >> >> IMO, there are some aspects, especially those at the boundary of Hive and >> Flink, that need a bit more discussion. For example >> >> * What does it take to make Flink schema compatible with Hive schema? >> * How will Flink tables (descriptors) be stored in HMS? >> * How do both Hive catalogs differ? Could they be integrated into to a >> single one? When to use which one? >> * What meta information is provided by HMS? What of this can be leveraged >> by Flink? >> >> Thank you, >> Fabian >> >> Am Fr., 2. Nov. 2018 um 00:31 Uhr schrieb Bowen Li <[hidden email]>: >> >>> After taking a look at how other discussion threads work, I think it's >>> actually fine just keep our discussion here. It's up to you, Xuefu. >>> >>> The google doc LGTM. I left some minor comments. >>> >>> On Thu, Nov 1, 2018 at 10:17 AM Bowen Li <[hidden email]> wrote: >>> >>>> Hi all, >>>> >>>> As Xuefu has published the design doc on google, I agree with Shuyi's >>>> suggestion that we probably should start a new email thread like "[DISCUSS] >>>> ... Hive integration design ..." on only dev mailing list for community >>>> devs to review. The current thread sends to both dev and user list. >>>> >>>> This email thread is more like validating the general idea and direction >>>> with the community, and it's been pretty long and crowded so far. Since >>>> everyone is pro for the idea, we can move forward with another thread to >>>> discuss and finalize the design. >>>> >>>> Thanks, >>>> Bowen >>>> >>>> On Wed, Oct 31, 2018 at 12:16 PM Zhang, Xuefu <[hidden email]> >>>> wrote: >>>> >>>>> Hi Shuiyi, >>>>> >>>>> Good idea. Actually the PDF was converted from a google doc. Here is its >>>>> link: >>>>> >>>>> https://docs.google.com/document/d/1SkppRD_rE3uOKSN-LuZCqn4f7dz0zW5aa6T_hBZq5_o/edit?usp=sharing >>>>> Once we reach an agreement, I can convert it to a FLIP. >>>>> >>>>> Thanks, >>>>> Xuefu >>>>> >>>>> >>>>> >>>>> ------------------------------------------------------------------ >>>>> Sender:Shuyi Chen <[hidden email]> >>>>> Sent at:2018 Nov 1 (Thu) 02:47 >>>>> Recipient:Xuefu <[hidden email]> >>>>> Cc:vino yang <[hidden email]>; Fabian Hueske <[hidden email]>; >>>>> dev <[hidden email]>; user <[hidden email]> >>>>> Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem >>>>> >>>>> Hi Xuefu, >>>>> >>>>> Thanks a lot for driving this big effort. I would suggest convert your >>>>> proposal and design doc into a google doc, and share it on the dev mailing >>>>> list for the community to review and comment with title like "[DISCUSS] ... >>>>> Hive integration design ..." . Once approved, we can document it as a FLIP >>>>> (Flink Improvement Proposals), and use JIRAs to track the implementations. >>>>> What do you think? >>>>> >>>>> Shuyi >>>>> >>>>> On Tue, Oct 30, 2018 at 11:32 AM Zhang, Xuefu <[hidden email]> >>>>> wrote: >>>>> Hi all, >>>>> >>>>> I have also shared a design doc on Hive metastore integration that is >>>>> attached here and also to FLINK-10556[1]. Please kindly review and share >>>>> your feedback. >>>>> >>>>> >>>>> Thanks, >>>>> Xuefu >>>>> >>>>> [1] https://issues.apache.org/jira/browse/FLINK-10556 >>>>> ------------------------------------------------------------------ >>>>> Sender:Xuefu <[hidden email]> >>>>> Sent at:2018 Oct 25 (Thu) 01:08 >>>>> Recipient:Xuefu <[hidden email]>; Shuyi Chen < >>>>> [hidden email]> >>>>> Cc:yanghua1127 <[hidden email]>; Fabian Hueske <[hidden email]>; >>>>> dev <[hidden email]>; user <[hidden email]> >>>>> Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem >>>>> >>>>> Hi all, >>>>> >>>>> To wrap up the discussion, I have attached a PDF describing the >>>>> proposal, which is also attached to FLINK-10556 [1]. Please feel free to >>>>> watch that JIRA to track the progress. >>>>> >>>>> Please also let me know if you have additional comments or questions. >>>>> >>>>> Thanks, >>>>> Xuefu >>>>> >>>>> [1] https://issues.apache.org/jira/browse/FLINK-10556 >>>>> >>>>> >>>>> ------------------------------------------------------------------ >>>>> Sender:Xuefu <[hidden email]> >>>>> Sent at:2018 Oct 16 (Tue) 03:40 >>>>> Recipient:Shuyi Chen <[hidden email]> >>>>> Cc:yanghua1127 <[hidden email]>; Fabian Hueske <[hidden email]>; >>>>> dev <[hidden email]>; user <[hidden email]> >>>>> Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem >>>>> >>>>> Hi Shuyi, >>>>> >>>>> Thank you for your input. Yes, I agreed with a phased approach and like >>>>> to move forward fast. :) We did some work internally on DDL utilizing babel >>>>> parser in Calcite. While babel makes Calcite's grammar extensible, at >>>>> first impression it still seems too cumbersome for a project when too >>>>> much extensions are made. It's even challenging to find where the extension >>>>> is needed! It would be certainly better if Calcite can magically support >>>>> Hive QL by just turning on a flag, such as that for MYSQL_5. I can also >>>>> see that this could mean a lot of work on Calcite. Nevertheless, I will >>>>> bring up the discussion over there and to see what their community thinks. >>>>> >>>>> Would mind to share more info about the proposal on DDL that you >>>>> mentioned? We can certainly collaborate on this. >>>>> >>>>> Thanks, >>>>> Xuefu >>>>> >>>>> ------------------------------------------------------------------ >>>>> Sender:Shuyi Chen <[hidden email]> >>>>> Sent at:2018 Oct 14 (Sun) 08:30 >>>>> Recipient:Xuefu <[hidden email]> >>>>> Cc:yanghua1127 <[hidden email]>; Fabian Hueske <[hidden email]>; >>>>> dev <[hidden email]>; user <[hidden email]> >>>>> Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem >>>>> >>>>> Welcome to the community and thanks for the great proposal, Xuefu! I >>>>> think the proposal can be divided into 2 stages: making Flink to support >>>>> Hive features, and make Hive to work with Flink. I agreed with Timo that on >>>>> starting with a smaller scope, so we can make progress faster. As for [6], >>>>> a proposal for DDL is already in progress, and will come after the unified >>>>> SQL connector API is done. For supporting Hive syntax, we might need to >>>>> work with the Calcite community, and a recent effort called babel ( >>>>> https://issues.apache.org/jira/browse/CALCITE-2280) in Calcite might >>>>> help here. >>>>> >>>>> Thanks >>>>> Shuyi >>>>> >>>>> On Wed, Oct 10, 2018 at 8:02 PM Zhang, Xuefu <[hidden email]> >>>>> wrote: >>>>> Hi Fabian/Vno, >>>>> >>>>> Thank you very much for your encouragement inquiry. Sorry that I didn't >>>>> see Fabian's email until I read Vino's response just now. (Somehow Fabian's >>>>> went to the spam folder.) >>>>> >>>>> My proposal contains long-term and short-terms goals. Nevertheless, the >>>>> effort will focus on the following areas, including Fabian's list: >>>>> >>>>> 1. Hive metastore connectivity - This covers both read/write access, >>>>> which means Flink can make full use of Hive's metastore as its catalog (at >>>>> least for the batch but can extend for streaming as well). >>>>> 2. Metadata compatibility - Objects (databases, tables, partitions, etc) >>>>> created by Hive can be understood by Flink and the reverse direction is >>>>> true also. >>>>> 3. Data compatibility - Similar to #2, data produced by Hive can be >>>>> consumed by Flink and vise versa. >>>>> 4. Support Hive UDFs - For all Hive's native udfs, Flink either provides >>>>> its own implementation or make Hive's implementation work in Flink. >>>>> Further, for user created UDFs in Hive, Flink SQL should provide a >>>>> mechanism allowing user to import them into Flink without any code change >>>>> required. >>>>> 5. Data types - Flink SQL should support all data types that are >>>>> available in Hive. >>>>> 6. SQL Language - Flink SQL should support SQL standard (such as >>>>> SQL2003) with extension to support Hive's syntax and language features, >>>>> around DDL, DML, and SELECT queries. >>>>> 7. SQL CLI - this is currently developing in Flink but more effort is >>>>> needed. >>>>> 8. Server - provide a server that's compatible with Hive's HiverServer2 >>>>> in thrift APIs, such that HiveServer2 users can reuse their existing client >>>>> (such as beeline) but connect to Flink's thrift server instead. >>>>> 9. JDBC/ODBC drivers - Flink may provide its own JDBC/ODBC drivers for >>>>> other application to use to connect to its thrift server >>>>> 10. Support other user's customizations in Hive, such as Hive Serdes, >>>>> storage handlers, etc. >>>>> 11. Better task failure tolerance and task scheduling at Flink runtime. >>>>> >>>>> As you can see, achieving all those requires significant effort and >>>>> across all layers in Flink. However, a short-term goal could include only >>>>> core areas (such as 1, 2, 4, 5, 6, 7) or start at a smaller scope (such as >>>>> #3, #6). >>>>> >>>>> Please share your further thoughts. If we generally agree that this is >>>>> the right direction, I could come up with a formal proposal quickly and >>>>> then we can follow up with broader discussions. >>>>> >>>>> Thanks, >>>>> Xuefu >>>>> >>>>> >>>>> >>>>> ------------------------------------------------------------------ >>>>> Sender:vino yang <[hidden email]> >>>>> Sent at:2018 Oct 11 (Thu) 09:45 >>>>> Recipient:Fabian Hueske <[hidden email]> >>>>> Cc:dev <[hidden email]>; Xuefu <[hidden email]>; user < >>>>> [hidden email]> >>>>> Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem >>>>> >>>>> Hi Xuefu, >>>>> >>>>> Appreciate this proposal, and like Fabian, it would look better if you >>>>> can give more details of the plan. >>>>> >>>>> Thanks, vino. >>>>> >>>>> Fabian Hueske <[hidden email]> 于2018年10月10日周三 下午5:27写道： >>>>> Hi Xuefu, >>>>> >>>>> Welcome to the Flink community and thanks for starting this discussion! >>>>> Better Hive integration would be really great! >>>>> Can you go into details of what you are proposing? I can think of a >>>>> couple ways to improve Flink in that regard: >>>>> >>>>> * Support for Hive UDFs >>>>> * Support for Hive metadata catalog >>>>> * Support for HiveQL syntax >>>>> * ??? >>>>> >>>>> Best, Fabian >>>>> >>>>> Am Di., 9. Okt. 2018 um 19:22 Uhr schrieb Zhang, Xuefu < >>>>> [hidden email]>: >>>>> Hi all, >>>>> >>>>> Along with the community's effort, inside Alibaba we have explored >>>>> Flink's potential as an execution engine not just for stream processing but >>>>> also for batch processing. We are encouraged by our findings and have >>>>> initiated our effort to make Flink's SQL capabilities full-fledged. When >>>>> comparing what's available in Flink to the offerings from competitive data >>>>> processing engines, we identified a major gap in Flink: a well integration >>>>> with Hive ecosystem. This is crucial to the success of Flink SQL and batch >>>>> due to the well-established data ecosystem around Hive. Therefore, we have >>>>> done some initial work along this direction but there are still a lot of >>>>> effort needed. >>>>> >>>>> We have two strategies in mind. The first one is to make Flink SQL >>>>> full-fledged and well-integrated with Hive ecosystem. This is a similar >>>>> approach to what Spark SQL adopted. The second strategy is to make Hive >>>>> itself work with Flink, similar to the proposal in [1]. Each approach bears >>>>> its pros and cons, but they don’t need to be mutually exclusive with each >>>>> targeting at different users and use cases. We believe that both will >>>>> promote a much greater adoption of Flink beyond stream processing. >>>>> >>>>> We have been focused on the first approach and would like to showcase >>>>> Flink's batch and SQL capabilities with Flink SQL. However, we have also >>>>> planned to start strategy #2 as the follow-up effort. >>>>> >>>>> I'm completely new to Flink(, with a short bio [2] below), though many >>>>> of my colleagues here at Alibaba are long-time contributors. Nevertheless, >>>>> I'd like to share our thoughts and invite your early feedback. At the same >>>>> time, I am working on a detailed proposal on Flink SQL's integration with >>>>> Hive ecosystem, which will be also shared when ready. >>>>> >>>>> While the ideas are simple, each approach will demand significant >>>>> effort, more than what we can afford. Thus, the input and contributions >>>>> from the communities are greatly welcome and appreciated. >>>>> >>>>> Regards, >>>>> >>>>> >>>>> Xuefu >>>>> >>>>> References: >>>>> >>>>> [1] https://issues.apache.org/jira/browse/HIVE-10712 >>>>> [2] Xuefu Zhang is a long-time open source veteran, worked or working on >>>>> many projects under Apache Foundation, of which he is also an honored >>>>> member. About 10 years ago he worked in the Hadoop team at Yahoo where the >>>>> projects just got started. Later he worked at Cloudera, initiating and >>>>> leading the development of Hive on Spark project in the communities and >>>>> across many organizations. Prior to joining Alibaba, he worked at Uber >>>>> where he promoted Hive on Spark to all Uber's SQL on Hadoop workload and >>>>> significantly improved Uber's cluster efficiency. >>>>> >>>>> >>>>> >>>>> >>>>> -- >>>>> "So you have to trust that the dots will somehow connect in your future." >>>>> >>>>> >>>>> -- >>>>> "So you have to trust that the dots will somehow connect in your future." >>>>> |
Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem
|
|
Hi Piotr,
That sounds good to me. Let's close all the open questions ((there are a couple of them)) in the Google doc and I should be able to quickly split it into the three proposals as you suggested. Thanks, Xuefu ------------------------------------------------------------------ Sender:Piotr Nowojski <[hidden email]> Sent at:2018 Nov 9 (Fri) 22:46 Recipient:dev <[hidden email]>; Xuefu <[hidden email]> Cc:Bowen Li <[hidden email]>; Shuyi Chen <[hidden email]> Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem Hi, Yes, it seems like the best solution. Maybe someone else can also suggests if we can split it further? Maybe changes in the interface in one doc, reading from hive meta store another and final storing our meta informations in hive meta store? Piotrek > On 9 Nov 2018, at 01:44, Zhang, Xuefu <[hidden email]> wrote: > > Hi Piotr, > > That seems to be good idea! > > Since the google doc for the design is currently under extensive review, I will leave it as it is for now. However, I'll convert it to two different FLIPs when the time comes. > > How does it sound to you? > > Thanks, > Xuefu > > > ------------------------------------------------------------------ > Sender:Piotr Nowojski <[hidden email]> > Sent at:2018 Nov 9 (Fri) 02:31 > Recipient:dev <[hidden email]> > Cc:Bowen Li <[hidden email]>; Xuefu <[hidden email]>; Shuyi Chen <[hidden email]> > Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem > > Hi, > > Maybe we should split this topic (and the design doc) into couple of smaller ones, hopefully independent. The questions that you have asked Fabian have for example very little to do with reading metadata from Hive Meta Store? > > Piotrek > >> On 7 Nov 2018, at 14:27, Fabian Hueske <[hidden email]> wrote: >> >> Hi Xuefu and all, >> >> Thanks for sharing this design document! >> I'm very much in favor of restructuring / reworking the catalog handling in >> Flink SQL as outlined in the document. >> Most changes described in the design document seem to be rather general and >> not specifically related to the Hive integration. >> >> IMO, there are some aspects, especially those at the boundary of Hive and >> Flink, that need a bit more discussion. For example >> >> * What does it take to make Flink schema compatible with Hive schema? >> * How will Flink tables (descriptors) be stored in HMS? >> * How do both Hive catalogs differ? Could they be integrated into to a >> single one? When to use which one? >> * What meta information is provided by HMS? What of this can be leveraged >> by Flink? >> >> Thank you, >> Fabian >> >> Am Fr., 2. Nov. 2018 um 00:31 Uhr schrieb Bowen Li <[hidden email]>: >> >>> After taking a look at how other discussion threads work, I think it's >>> actually fine just keep our discussion here. It's up to you, Xuefu. >>> >>> The google doc LGTM. I left some minor comments. >>> >>> On Thu, Nov 1, 2018 at 10:17 AM Bowen Li <[hidden email]> wrote: >>> >>>> Hi all, >>>> >>>> As Xuefu has published the design doc on google, I agree with Shuyi's >>>> suggestion that we probably should start a new email thread like "[DISCUSS] >>>> ... Hive integration design ..." on only dev mailing list for community >>>> devs to review. The current thread sends to both dev and user list. >>>> >>>> This email thread is more like validating the general idea and direction >>>> with the community, and it's been pretty long and crowded so far. Since >>>> everyone is pro for the idea, we can move forward with another thread to >>>> discuss and finalize the design. >>>> >>>> Thanks, >>>> Bowen >>>> >>>> On Wed, Oct 31, 2018 at 12:16 PM Zhang, Xuefu <[hidden email]> >>>> wrote: >>>> >>>>> Hi Shuiyi, >>>>> >>>>> Good idea. Actually the PDF was converted from a google doc. Here is its >>>>> link: >>>>> >>>>> https://docs.google.com/document/d/1SkppRD_rE3uOKSN-LuZCqn4f7dz0zW5aa6T_hBZq5_o/edit?usp=sharing >>>>> Once we reach an agreement, I can convert it to a FLIP. >>>>> >>>>> Thanks, >>>>> Xuefu >>>>> >>>>> >>>>> >>>>> ------------------------------------------------------------------ >>>>> Sender:Shuyi Chen <[hidden email]> >>>>> Sent at:2018 Nov 1 (Thu) 02:47 >>>>> Recipient:Xuefu <[hidden email]> >>>>> Cc:vino yang <[hidden email]>; Fabian Hueske <[hidden email]>; >>>>> dev <[hidden email]>; user <[hidden email]> >>>>> Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem >>>>> >>>>> Hi Xuefu, >>>>> >>>>> Thanks a lot for driving this big effort. I would suggest convert your >>>>> proposal and design doc into a google doc, and share it on the dev mailing >>>>> list for the community to review and comment with title like "[DISCUSS] ... >>>>> Hive integration design ..." . Once approved, we can document it as a FLIP >>>>> (Flink Improvement Proposals), and use JIRAs to track the implementations. >>>>> What do you think? >>>>> >>>>> Shuyi >>>>> >>>>> On Tue, Oct 30, 2018 at 11:32 AM Zhang, Xuefu <[hidden email]> >>>>> wrote: >>>>> Hi all, >>>>> >>>>> I have also shared a design doc on Hive metastore integration that is >>>>> attached here and also to FLINK-10556[1]. Please kindly review and share >>>>> your feedback. >>>>> >>>>> >>>>> Thanks, >>>>> Xuefu >>>>> >>>>> [1] https://issues.apache.org/jira/browse/FLINK-10556 >>>>> ------------------------------------------------------------------ >>>>> Sender:Xuefu <[hidden email]> >>>>> Sent at:2018 Oct 25 (Thu) 01:08 >>>>> Recipient:Xuefu <[hidden email]>; Shuyi Chen < >>>>> [hidden email]> >>>>> Cc:yanghua1127 <[hidden email]>; Fabian Hueske <[hidden email]>; >>>>> dev <[hidden email]>; user <[hidden email]> >>>>> Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem >>>>> >>>>> Hi all, >>>>> >>>>> To wrap up the discussion, I have attached a PDF describing the >>>>> proposal, which is also attached to FLINK-10556 [1]. Please feel free to >>>>> watch that JIRA to track the progress. >>>>> >>>>> Please also let me know if you have additional comments or questions. >>>>> >>>>> Thanks, >>>>> Xuefu >>>>> >>>>> [1] https://issues.apache.org/jira/browse/FLINK-10556 >>>>> >>>>> >>>>> ------------------------------------------------------------------ >>>>> Sender:Xuefu <[hidden email]> >>>>> Sent at:2018 Oct 16 (Tue) 03:40 >>>>> Recipient:Shuyi Chen <[hidden email]> >>>>> Cc:yanghua1127 <[hidden email]>; Fabian Hueske <[hidden email]>; >>>>> dev <[hidden email]>; user <[hidden email]> >>>>> Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem >>>>> >>>>> Hi Shuyi, >>>>> >>>>> Thank you for your input. Yes, I agreed with a phased approach and like >>>>> to move forward fast. :) We did some work internally on DDL utilizing babel >>>>> parser in Calcite. While babel makes Calcite's grammar extensible, at >>>>> first impression it still seems too cumbersome for a project when too >>>>> much extensions are made. It's even challenging to find where the extension >>>>> is needed! It would be certainly better if Calcite can magically support >>>>> Hive QL by just turning on a flag, such as that for MYSQL_5. I can also >>>>> see that this could mean a lot of work on Calcite. Nevertheless, I will >>>>> bring up the discussion over there and to see what their community thinks. >>>>> >>>>> Would mind to share more info about the proposal on DDL that you >>>>> mentioned? We can certainly collaborate on this. >>>>> >>>>> Thanks, >>>>> Xuefu >>>>> >>>>> ------------------------------------------------------------------ >>>>> Sender:Shuyi Chen <[hidden email]> >>>>> Sent at:2018 Oct 14 (Sun) 08:30 >>>>> Recipient:Xuefu <[hidden email]> >>>>> Cc:yanghua1127 <[hidden email]>; Fabian Hueske <[hidden email]>; >>>>> dev <[hidden email]>; user <[hidden email]> >>>>> Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem >>>>> >>>>> Welcome to the community and thanks for the great proposal, Xuefu! I >>>>> think the proposal can be divided into 2 stages: making Flink to support >>>>> Hive features, and make Hive to work with Flink. I agreed with Timo that on >>>>> starting with a smaller scope, so we can make progress faster. As for [6], >>>>> a proposal for DDL is already in progress, and will come after the unified >>>>> SQL connector API is done. For supporting Hive syntax, we might need to >>>>> work with the Calcite community, and a recent effort called babel ( >>>>> https://issues.apache.org/jira/browse/CALCITE-2280) in Calcite might >>>>> help here. >>>>> >>>>> Thanks >>>>> Shuyi >>>>> >>>>> On Wed, Oct 10, 2018 at 8:02 PM Zhang, Xuefu <[hidden email]> >>>>> wrote: >>>>> Hi Fabian/Vno, >>>>> >>>>> Thank you very much for your encouragement inquiry. Sorry that I didn't >>>>> see Fabian's email until I read Vino's response just now. (Somehow Fabian's >>>>> went to the spam folder.) >>>>> >>>>> My proposal contains long-term and short-terms goals. Nevertheless, the >>>>> effort will focus on the following areas, including Fabian's list: >>>>> >>>>> 1. Hive metastore connectivity - This covers both read/write access, >>>>> which means Flink can make full use of Hive's metastore as its catalog (at >>>>> least for the batch but can extend for streaming as well). >>>>> 2. Metadata compatibility - Objects (databases, tables, partitions, etc) >>>>> created by Hive can be understood by Flink and the reverse direction is >>>>> true also. >>>>> 3. Data compatibility - Similar to #2, data produced by Hive can be >>>>> consumed by Flink and vise versa. >>>>> 4. Support Hive UDFs - For all Hive's native udfs, Flink either provides >>>>> its own implementation or make Hive's implementation work in Flink. >>>>> Further, for user created UDFs in Hive, Flink SQL should provide a >>>>> mechanism allowing user to import them into Flink without any code change >>>>> required. >>>>> 5. Data types - Flink SQL should support all data types that are >>>>> available in Hive. >>>>> 6. SQL Language - Flink SQL should support SQL standard (such as >>>>> SQL2003) with extension to support Hive's syntax and language features, >>>>> around DDL, DML, and SELECT queries. >>>>> 7. SQL CLI - this is currently developing in Flink but more effort is >>>>> needed. >>>>> 8. Server - provide a server that's compatible with Hive's HiverServer2 >>>>> in thrift APIs, such that HiveServer2 users can reuse their existing client >>>>> (such as beeline) but connect to Flink's thrift server instead. >>>>> 9. JDBC/ODBC drivers - Flink may provide its own JDBC/ODBC drivers for >>>>> other application to use to connect to its thrift server >>>>> 10. Support other user's customizations in Hive, such as Hive Serdes, >>>>> storage handlers, etc. >>>>> 11. Better task failure tolerance and task scheduling at Flink runtime. >>>>> >>>>> As you can see, achieving all those requires significant effort and >>>>> across all layers in Flink. However, a short-term goal could include only >>>>> core areas (such as 1, 2, 4, 5, 6, 7) or start at a smaller scope (such as >>>>> #3, #6). >>>>> >>>>> Please share your further thoughts. If we generally agree that this is >>>>> the right direction, I could come up with a formal proposal quickly and >>>>> then we can follow up with broader discussions. >>>>> >>>>> Thanks, >>>>> Xuefu >>>>> >>>>> >>>>> >>>>> ------------------------------------------------------------------ >>>>> Sender:vino yang <[hidden email]> >>>>> Sent at:2018 Oct 11 (Thu) 09:45 >>>>> Recipient:Fabian Hueske <[hidden email]> >>>>> Cc:dev <[hidden email]>; Xuefu <[hidden email]>; user < >>>>> [hidden email]> >>>>> Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem >>>>> >>>>> Hi Xuefu, >>>>> >>>>> Appreciate this proposal, and like Fabian, it would look better if you >>>>> can give more details of the plan. >>>>> >>>>> Thanks, vino. >>>>> >>>>> Fabian Hueske <[hidden email]> 于2018年10月10日周三 下午5:27写道： >>>>> Hi Xuefu, >>>>> >>>>> Welcome to the Flink community and thanks for starting this discussion! >>>>> Better Hive integration would be really great! >>>>> Can you go into details of what you are proposing? I can think of a >>>>> couple ways to improve Flink in that regard: >>>>> >>>>> * Support for Hive UDFs >>>>> * Support for Hive metadata catalog >>>>> * Support for HiveQL syntax >>>>> * ??? >>>>> >>>>> Best, Fabian >>>>> >>>>> Am Di., 9. Okt. 2018 um 19:22 Uhr schrieb Zhang, Xuefu < >>>>> [hidden email]>: >>>>> Hi all, >>>>> >>>>> Along with the community's effort, inside Alibaba we have explored >>>>> Flink's potential as an execution engine not just for stream processing but >>>>> also for batch processing. We are encouraged by our findings and have >>>>> initiated our effort to make Flink's SQL capabilities full-fledged. When >>>>> comparing what's available in Flink to the offerings from competitive data >>>>> processing engines, we identified a major gap in Flink: a well integration >>>>> with Hive ecosystem. This is crucial to the success of Flink SQL and batch >>>>> due to the well-established data ecosystem around Hive. Therefore, we have >>>>> done some initial work along this direction but there are still a lot of >>>>> effort needed. >>>>> >>>>> We have two strategies in mind. The first one is to make Flink SQL >>>>> full-fledged and well-integrated with Hive ecosystem. This is a similar >>>>> approach to what Spark SQL adopted. The second strategy is to make Hive >>>>> itself work with Flink, similar to the proposal in [1]. Each approach bears >>>>> its pros and cons, but they don’t need to be mutually exclusive with each >>>>> targeting at different users and use cases. We believe that both will >>>>> promote a much greater adoption of Flink beyond stream processing. >>>>> >>>>> We have been focused on the first approach and would like to showcase >>>>> Flink's batch and SQL capabilities with Flink SQL. However, we have also >>>>> planned to start strategy #2 as the follow-up effort. >>>>> >>>>> I'm completely new to Flink(, with a short bio [2] below), though many >>>>> of my colleagues here at Alibaba are long-time contributors. Nevertheless, >>>>> I'd like to share our thoughts and invite your early feedback. At the same >>>>> time, I am working on a detailed proposal on Flink SQL's integration with >>>>> Hive ecosystem, which will be also shared when ready. >>>>> >>>>> While the ideas are simple, each approach will demand significant >>>>> effort, more than what we can afford. Thus, the input and contributions >>>>> from the communities are greatly welcome and appreciated. >>>>> >>>>> Regards, >>>>> >>>>> >>>>> Xuefu >>>>> >>>>> References: >>>>> >>>>> [1] https://issues.apache.org/jira/browse/HIVE-10712 >>>>> [2] Xuefu Zhang is a long-time open source veteran, worked or working on >>>>> many projects under Apache Foundation, of which he is also an honored >>>>> member. About 10 years ago he worked in the Hadoop team at Yahoo where the >>>>> projects just got started. Later he worked at Cloudera, initiating and >>>>> leading the development of Hive on Spark project in the communities and >>>>> across many organizations. Prior to joining Alibaba, he worked at Uber >>>>> where he promoted Hive on Spark to all Uber's SQL on Hadoop workload and >>>>> significantly improved Uber's cluster efficiency. >>>>> >>>>> >>>>> >>>>> >>>>> -- >>>>> "So you have to trust that the dots will somehow connect in your future." >>>>> >>>>> >>>>> -- >>>>> "So you have to trust that the dots will somehow connect in your future." >>>>> |
Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem
|
|
Hi Piotr
I have extracted the API portion of the design and the google doc is here. Please review and provide your feedback. Thanks, Xuefu ------------------------------------------------------------------ Sender:Xuefu <[hidden email]> Sent at:2018 Nov 12 (Mon) 12:43 Recipient:Piotr Nowojski <[hidden email]>; dev <[hidden email]> Cc:Bowen Li <[hidden email]>; Shuyi Chen <[hidden email]> Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem Hi Piotr, That sounds good to me. Let's close all the open questions ((there are a couple of them)) in the Google doc and I should be able to quickly split it into the three proposals as you suggested. Thanks, Xuefu ------------------------------------------------------------------ Sender:Piotr Nowojski <[hidden email]> Sent at:2018 Nov 9 (Fri) 22:46 Recipient:dev <[hidden email]>; Xuefu <[hidden email]> Cc:Bowen Li <[hidden email]>; Shuyi Chen <[hidden email]> Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem Hi, Yes, it seems like the best solution. Maybe someone else can also suggests if we can split it further? Maybe changes in the interface in one doc, reading from hive meta store another and final storing our meta informations in hive meta store? Piotrek > On 9 Nov 2018, at 01:44, Zhang, Xuefu <[hidden email]> wrote: > > Hi Piotr, > > That seems to be good idea! > > Since the google doc for the design is currently under extensive review, I will leave it as it is for now. However, I'll convert it to two different FLIPs when the time comes. > > How does it sound to you? > > Thanks, > Xuefu > > > ------------------------------------------------------------------ > Sender:Piotr Nowojski <[hidden email]> > Sent at:2018 Nov 9 (Fri) 02:31 > Recipient:dev <[hidden email]> > Cc:Bowen Li <[hidden email]>; Xuefu <[hidden email]>; Shuyi Chen <[hidden email]> > Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem > > Hi, > > Maybe we should split this topic (and the design doc) into couple of smaller ones, hopefully independent. The questions that you have asked Fabian have for example very little to do with reading metadata from Hive Meta Store? > > Piotrek > >> On 7 Nov 2018, at 14:27, Fabian Hueske <[hidden email]> wrote: >> >> Hi Xuefu and all, >> >> Thanks for sharing this design document! >> I'm very much in favor of restructuring / reworking the catalog handling in >> Flink SQL as outlined in the document. >> Most changes described in the design document seem to be rather general and >> not specifically related to the Hive integration. >> >> IMO, there are some aspects, especially those at the boundary of Hive and >> Flink, that need a bit more discussion. For example >> >> * What does it take to make Flink schema compatible with Hive schema? >> * How will Flink tables (descriptors) be stored in HMS? >> * How do both Hive catalogs differ? Could they be integrated into to a >> single one? When to use which one? >> * What meta information is provided by HMS? What of this can be leveraged >> by Flink? >> >> Thank you, >> Fabian >> >> Am Fr., 2. Nov. 2018 um 00:31 Uhr schrieb Bowen Li <[hidden email]>: >> >>> After taking a look at how other discussion threads work, I think it's >>> actually fine just keep our discussion here. It's up to you, Xuefu. >>> >>> The google doc LGTM. I left some minor comments. >>> >>> On Thu, Nov 1, 2018 at 10:17 AM Bowen Li <[hidden email]> wrote: >>> >>>> Hi all, >>>> >>>> As Xuefu has published the design doc on google, I agree with Shuyi's >>>> suggestion that we probably should start a new email thread like "[DISCUSS] >>>> ... Hive integration design ..." on only dev mailing list for community >>>> devs to review. The current thread sends to both dev and user list. >>>> >>>> This email thread is more like validating the general idea and direction >>>> with the community, and it's been pretty long and crowded so far. Since >>>> everyone is pro for the idea, we can move forward with another thread to >>>> discuss and finalize the design. >>>> >>>> Thanks, >>>> Bowen >>>> >>>> On Wed, Oct 31, 2018 at 12:16 PM Zhang, Xuefu <[hidden email]> >>>> wrote: >>>> >>>>> Hi Shuiyi, >>>>> >>>>> Good idea. Actually the PDF was converted from a google doc. Here is its >>>>> link: >>>>> >>>>> https://docs.google.com/document/d/1SkppRD_rE3uOKSN-LuZCqn4f7dz0zW5aa6T_hBZq5_o/edit?usp=sharing >>>>> Once we reach an agreement, I can convert it to a FLIP. >>>>> >>>>> Thanks, >>>>> Xuefu >>>>> >>>>> >>>>> >>>>> ------------------------------------------------------------------ >>>>> Sender:Shuyi Chen <[hidden email]> >>>>> Sent at:2018 Nov 1 (Thu) 02:47 >>>>> Recipient:Xuefu <[hidden email]> >>>>> Cc:vino yang <[hidden email]>; Fabian Hueske <[hidden email]>; >>>>> dev <[hidden email]>; user <[hidden email]> >>>>> Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem >>>>> >>>>> Hi Xuefu, >>>>> >>>>> Thanks a lot for driving this big effort. I would suggest convert your >>>>> proposal and design doc into a google doc, and share it on the dev mailing >>>>> list for the community to review and comment with title like "[DISCUSS] ... >>>>> Hive integration design ..." . Once approved, we can document it as a FLIP >>>>> (Flink Improvement Proposals), and use JIRAs to track the implementations. >>>>> What do you think? >>>>> >>>>> Shuyi >>>>> >>>>> On Tue, Oct 30, 2018 at 11:32 AM Zhang, Xuefu <[hidden email]> >>>>> wrote: >>>>> Hi all, >>>>> >>>>> I have also shared a design doc on Hive metastore integration that is >>>>> attached here and also to FLINK-10556[1]. Please kindly review and share >>>>> your feedback. >>>>> >>>>> >>>>> Thanks, >>>>> Xuefu >>>>> >>>>> [1] https://issues.apache.org/jira/browse/FLINK-10556 >>>>> ------------------------------------------------------------------ >>>>> Sender:Xuefu <[hidden email]> >>>>> Sent at:2018 Oct 25 (Thu) 01:08 >>>>> Recipient:Xuefu <[hidden email]>; Shuyi Chen < >>>>> [hidden email]> >>>>> Cc:yanghua1127 <[hidden email]>; Fabian Hueske <[hidden email]>; >>>>> dev <[hidden email]>; user <[hidden email]> >>>>> Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem >>>>> >>>>> Hi all, >>>>> >>>>> To wrap up the discussion, I have attached a PDF describing the >>>>> proposal, which is also attached to FLINK-10556 [1]. Please feel free to >>>>> watch that JIRA to track the progress. >>>>> >>>>> Please also let me know if you have additional comments or questions. >>>>> >>>>> Thanks, >>>>> Xuefu >>>>> >>>>> [1] https://issues.apache.org/jira/browse/FLINK-10556 >>>>> >>>>> >>>>> ------------------------------------------------------------------ >>>>> Sender:Xuefu <[hidden email]> >>>>> Sent at:2018 Oct 16 (Tue) 03:40 >>>>> Recipient:Shuyi Chen <[hidden email]> >>>>> Cc:yanghua1127 <[hidden email]>; Fabian Hueske <[hidden email]>; >>>>> dev <[hidden email]>; user <[hidden email]> >>>>> Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem >>>>> >>>>> Hi Shuyi, >>>>> >>>>> Thank you for your input. Yes, I agreed with a phased approach and like >>>>> to move forward fast. :) We did some work internally on DDL utilizing babel >>>>> parser in Calcite. While babel makes Calcite's grammar extensible, at >>>>> first impression it still seems too cumbersome for a project when too >>>>> much extensions are made. It's even challenging to find where the extension >>>>> is needed! It would be certainly better if Calcite can magically support >>>>> Hive QL by just turning on a flag, such as that for MYSQL_5. I can also >>>>> see that this could mean a lot of work on Calcite. Nevertheless, I will >>>>> bring up the discussion over there and to see what their community thinks. >>>>> >>>>> Would mind to share more info about the proposal on DDL that you >>>>> mentioned? We can certainly collaborate on this. >>>>> >>>>> Thanks, >>>>> Xuefu >>>>> >>>>> ------------------------------------------------------------------ >>>>> Sender:Shuyi Chen <[hidden email]> >>>>> Sent at:2018 Oct 14 (Sun) 08:30 >>>>> Recipient:Xuefu <[hidden email]> >>>>> Cc:yanghua1127 <[hidden email]>; Fabian Hueske <[hidden email]>; >>>>> dev <[hidden email]>; user <[hidden email]> >>>>> Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem >>>>> >>>>> Welcome to the community and thanks for the great proposal, Xuefu! I >>>>> think the proposal can be divided into 2 stages: making Flink to support >>>>> Hive features, and make Hive to work with Flink. I agreed with Timo that on >>>>> starting with a smaller scope, so we can make progress faster. As for [6], >>>>> a proposal for DDL is already in progress, and will come after the unified >>>>> SQL connector API is done. For supporting Hive syntax, we might need to >>>>> work with the Calcite community, and a recent effort called babel ( >>>>> https://issues.apache.org/jira/browse/CALCITE-2280) in Calcite might >>>>> help here. >>>>> >>>>> Thanks >>>>> Shuyi >>>>> >>>>> On Wed, Oct 10, 2018 at 8:02 PM Zhang, Xuefu <[hidden email]> >>>>> wrote: >>>>> Hi Fabian/Vno, >>>>> >>>>> Thank you very much for your encouragement inquiry. Sorry that I didn't >>>>> see Fabian's email until I read Vino's response just now. (Somehow Fabian's >>>>> went to the spam folder.) >>>>> >>>>> My proposal contains long-term and short-terms goals. Nevertheless, the >>>>> effort will focus on the following areas, including Fabian's list: >>>>> >>>>> 1. Hive metastore connectivity - This covers both read/write access, >>>>> which means Flink can make full use of Hive's metastore as its catalog (at >>>>> least for the batch but can extend for streaming as well). >>>>> 2. Metadata compatibility - Objects (databases, tables, partitions, etc) >>>>> created by Hive can be understood by Flink and the reverse direction is >>>>> true also. >>>>> 3. Data compatibility - Similar to #2, data produced by Hive can be >>>>> consumed by Flink and vise versa. >>>>> 4. Support Hive UDFs - For all Hive's native udfs, Flink either provides >>>>> its own implementation or make Hive's implementation work in Flink. >>>>> Further, for user created UDFs in Hive, Flink SQL should provide a >>>>> mechanism allowing user to import them into Flink without any code change >>>>> required. >>>>> 5. Data types - Flink SQL should support all data types that are >>>>> available in Hive. >>>>> 6. SQL Language - Flink SQL should support SQL standard (such as >>>>> SQL2003) with extension to support Hive's syntax and language features, >>>>> around DDL, DML, and SELECT queries. >>>>> 7. SQL CLI - this is currently developing in Flink but more effort is >>>>> needed. >>>>> 8. Server - provide a server that's compatible with Hive's HiverServer2 >>>>> in thrift APIs, such that HiveServer2 users can reuse their existing client >>>>> (such as beeline) but connect to Flink's thrift server instead. >>>>> 9. JDBC/ODBC drivers - Flink may provide its own JDBC/ODBC drivers for >>>>> other application to use to connect to its thrift server >>>>> 10. Support other user's customizations in Hive, such as Hive Serdes, >>>>> storage handlers, etc. >>>>> 11. Better task failure tolerance and task scheduling at Flink runtime. >>>>> >>>>> As you can see, achieving all those requires significant effort and >>>>> across all layers in Flink. However, a short-term goal could include only >>>>> core areas (such as 1, 2, 4, 5, 6, 7) or start at a smaller scope (such as >>>>> #3, #6). >>>>> >>>>> Please share your further thoughts. If we generally agree that this is >>>>> the right direction, I could come up with a formal proposal quickly and >>>>> then we can follow up with broader discussions. >>>>> >>>>> Thanks, >>>>> Xuefu >>>>> >>>>> >>>>> >>>>> ------------------------------------------------------------------ >>>>> Sender:vino yang <[hidden email]> >>>>> Sent at:2018 Oct 11 (Thu) 09:45 >>>>> Recipient:Fabian Hueske <[hidden email]> >>>>> Cc:dev <[hidden email]>; Xuefu <[hidden email]>; user < >>>>> [hidden email]> >>>>> Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem >>>>> >>>>> Hi Xuefu, >>>>> >>>>> Appreciate this proposal, and like Fabian, it would look better if you >>>>> can give more details of the plan. >>>>> >>>>> Thanks, vino. >>>>> >>>>> Fabian Hueske <[hidden email]> 于2018年10月10日周三 下午5:27写道： >>>>> Hi Xuefu, >>>>> >>>>> Welcome to the Flink community and thanks for starting this discussion! >>>>> Better Hive integration would be really great! >>>>> Can you go into details of what you are proposing? I can think of a >>>>> couple ways to improve Flink in that regard: >>>>> >>>>> * Support for Hive UDFs >>>>> * Support for Hive metadata catalog >>>>> * Support for HiveQL syntax >>>>> * ??? >>>>> >>>>> Best, Fabian >>>>> >>>>> Am Di., 9. Okt. 2018 um 19:22 Uhr schrieb Zhang, Xuefu < >>>>> [hidden email]>: >>>>> Hi all, >>>>> >>>>> Along with the community's effort, inside Alibaba we have explored >>>>> Flink's potential as an execution engine not just for stream processing but >>>>> also for batch processing. We are encouraged by our findings and have >>>>> initiated our effort to make Flink's SQL capabilities full-fledged. When >>>>> comparing what's available in Flink to the offerings from competitive data >>>>> processing engines, we identified a major gap in Flink: a well integration >>>>> with Hive ecosystem. This is crucial to the success of Flink SQL and batch >>>>> due to the well-established data ecosystem around Hive. Therefore, we have >>>>> done some initial work along this direction but there are still a lot of >>>>> effort needed. >>>>> >>>>> We have two strategies in mind. The first one is to make Flink SQL >>>>> full-fledged and well-integrated with Hive ecosystem. This is a similar >>>>> approach to what Spark SQL adopted. The second strategy is to make Hive >>>>> itself work with Flink, similar to the proposal in [1]. Each approach bears >>>>> its pros and cons, but they don’t need to be mutually exclusive with each >>>>> targeting at different users and use cases. We believe that both will >>>>> promote a much greater adoption of Flink beyond stream processing. >>>>> >>>>> We have been focused on the first approach and would like to showcase >>>>> Flink's batch and SQL capabilities with Flink SQL. However, we have also >>>>> planned to start strategy #2 as the follow-up effort. >>>>> >>>>> I'm completely new to Flink(, with a short bio [2] below), though many >>>>> of my colleagues here at Alibaba are long-time contributors. Nevertheless, >>>>> I'd like to share our thoughts and invite your early feedback. At the same >>>>> time, I am working on a detailed proposal on Flink SQL's integration with >>>>> Hive ecosystem, which will be also shared when ready. >>>>> >>>>> While the ideas are simple, each approach will demand significant >>>>> effort, more than what we can afford. Thus, the input and contributions >>>>> from the communities are greatly welcome and appreciated. >>>>> >>>>> Regards, >>>>> >>>>> >>>>> Xuefu >>>>> >>>>> References: >>>>> >>>>> [1] https://issues.apache.org/jira/browse/HIVE-10712 >>>>> [2] Xuefu Zhang is a long-time open source veteran, worked or working on >>>>> many projects under Apache Foundation, of which he is also an honored >>>>> member. About 10 years ago he worked in the Hadoop team at Yahoo where the >>>>> projects just got started. Later he worked at Cloudera, initiating and >>>>> leading the development of Hive on Spark project in the communities and >>>>> across many organizations. Prior to joining Alibaba, he worked at Uber >>>>> where he promoted Hive on Spark to all Uber's SQL on Hadoop workload and >>>>> significantly improved Uber's cluster efficiency. >>>>> >>>>> >>>>> >>>>> >>>>> -- >>>>> "So you have to trust that the dots will somehow connect in your future." >>>>> >>>>> >>>>> -- >>>>> "So you have to trust that the dots will somehow connect in your future." >>>>> |
|
|
Hi Xuefu,
Currently the new design doc <https://docs.google.com/document/d/1Y9it78yaUvbv4g572ZK_lZnZaAGjqwM_EhjdOv4yJtw/edit> is on “view only" mode, and people cannot leave comments. Can you please change it to "can comment" or "can edit" mode? Thanks, Bowen On Mon, Nov 12, 2018 at 9:51 PM Zhang, Xuefu <[hidden email]> wrote: > Hi Piotr > > I have extracted the API portion of the design and the google doc is here > <https://docs.google.com/document/d/1Y9it78yaUvbv4g572ZK_lZnZaAGjqwM_EhjdOv4yJtw/edit?usp=sharing>. > Please review and provide your feedback. > > Thanks, > Xuefu > > ------------------------------------------------------------------ > Sender:Xuefu <[hidden email]> > Sent at:2018 Nov 12 (Mon) 12:43 > Recipient:Piotr Nowojski <[hidden email]>; dev < > [hidden email]> > Cc:Bowen Li <[hidden email]>; Shuyi Chen <[hidden email]> > Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem > > Hi Piotr, > > That sounds good to me. Let's close all the open questions ((there are a > couple of them)) in the Google doc and I should be able to quickly split > it into the three proposals as you suggested. > > Thanks, > Xuefu > > ------------------------------------------------------------------ > Sender:Piotr Nowojski <[hidden email]> > Sent at:2018 Nov 9 (Fri) 22:46 > Recipient:dev <[hidden email]>; Xuefu <[hidden email]> > Cc:Bowen Li <[hidden email]>; Shuyi Chen <[hidden email]> > Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem > > Hi, > > > Yes, it seems like the best solution. Maybe someone else can also suggests if we can split it further? Maybe changes in the interface in one doc, reading from hive meta store another and final storing our meta informations in hive meta store? > > Piotrek > > > On 9 Nov 2018, at 01:44, Zhang, Xuefu <[hidden email]> wrote: > > > > Hi Piotr, > > > > That seems to be good idea! > > > > > Since the google doc for the design is currently under extensive review, I will leave it as it is for now. However, I'll convert it to two different FLIPs when the time comes. > > > > How does it sound to you? > > > > Thanks, > > Xuefu > > > > > > ------------------------------------------------------------------ > > Sender:Piotr Nowojski <[hidden email]> > > Sent at:2018 Nov 9 (Fri) 02:31 > > Recipient:dev <[hidden email]> > > Cc:Bowen Li <[hidden email]>; Xuefu <[hidden email] > >; Shuyi Chen <[hidden email]> > > Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem > > > > Hi, > > > > > Maybe we should split this topic (and the design doc) into couple of smaller ones, hopefully independent. The questions that you have asked Fabian have for example very little to do with reading metadata from Hive Meta Store? > > > > Piotrek > > > >> On 7 Nov 2018, at 14:27, Fabian Hueske <[hidden email]> wrote: > >> > >> Hi Xuefu and all, > >> > >> Thanks for sharing this design document! > > >> I'm very much in favor of restructuring / reworking the catalog handling in > >> Flink SQL as outlined in the document. > > >> Most changes described in the design document seem to be rather general and > >> not specifically related to the Hive integration. > >> > > >> IMO, there are some aspects, especially those at the boundary of Hive and > >> Flink, that need a bit more discussion. For example > >> > >> * What does it take to make Flink schema compatible with Hive schema? > >> * How will Flink tables (descriptors) be stored in HMS? > >> * How do both Hive catalogs differ? Could they be integrated into to a > >> single one? When to use which one? > > >> * What meta information is provided by HMS? What of this can be leveraged > >> by Flink? > >> > >> Thank you, > >> Fabian > >> > >> Am Fr., 2. Nov. 2018 um 00:31 Uhr schrieb Bowen Li <[hidden email] > >: > >> > >>> After taking a look at how other discussion threads work, I think it's > >>> actually fine just keep our discussion here. It's up to you, Xuefu. > >>> > >>> The google doc LGTM. I left some minor comments. > >>> > >>> On Thu, Nov 1, 2018 at 10:17 AM Bowen Li <[hidden email]> wrote: > >>> > >>>> Hi all, > >>>> > >>>> As Xuefu has published the design doc on google, I agree with Shuyi's > > >>>> suggestion that we probably should start a new email thread like "[DISCUSS] > > >>>> ... Hive integration design ..." on only dev mailing list for community > >>>> devs to review. The current thread sends to both dev and user list. > >>>> > > >>>> This email thread is more like validating the general idea and direction > > >>>> with the community, and it's been pretty long and crowded so far. Since > > >>>> everyone is pro for the idea, we can move forward with another thread to > >>>> discuss and finalize the design. > >>>> > >>>> Thanks, > >>>> Bowen > >>>> > >>>> On Wed, Oct 31, 2018 at 12:16 PM Zhang, Xuefu < > [hidden email]> > >>>> wrote: > >>>> > >>>>> Hi Shuiyi, > >>>>> > > >>>>> Good idea. Actually the PDF was converted from a google doc. Here is its > >>>>> link: > >>>>> > >>>>> > https://docs.google.com/document/d/1SkppRD_rE3uOKSN-LuZCqn4f7dz0zW5aa6T_hBZq5_o/edit?usp=sharing > >>>>> Once we reach an agreement, I can convert it to a FLIP. > >>>>> > >>>>> Thanks, > >>>>> Xuefu > >>>>> > >>>>> > >>>>> > >>>>> ------------------------------------------------------------------ > >>>>> Sender:Shuyi Chen <[hidden email]> > >>>>> Sent at:2018 Nov 1 (Thu) 02:47 > >>>>> Recipient:Xuefu <[hidden email]> > >>>>> Cc:vino yang <[hidden email]>; Fabian Hueske < > [hidden email]>; > >>>>> dev <[hidden email]>; user <[hidden email]> > >>>>> Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem > >>>>> > >>>>> Hi Xuefu, > >>>>> > > >>>>> Thanks a lot for driving this big effort. I would suggest convert your > > >>>>> proposal and design doc into a google doc, and share it on the dev mailing > > >>>>> list for the community to review and comment with title like "[DISCUSS] ... > > >>>>> Hive integration design ..." . Once approved, we can document it as a FLIP > > >>>>> (Flink Improvement Proposals), and use JIRAs to track the implementations. > >>>>> What do you think? > >>>>> > >>>>> Shuyi > >>>>> > >>>>> On Tue, Oct 30, 2018 at 11:32 AM Zhang, Xuefu < > [hidden email]> > >>>>> wrote: > >>>>> Hi all, > >>>>> > >>>>> I have also shared a design doc on Hive metastore integration that is > > >>>>> attached here and also to FLINK-10556[1]. Please kindly review and share > >>>>> your feedback. > >>>>> > >>>>> > >>>>> Thanks, > >>>>> Xuefu > >>>>> > >>>>> [1] https://issues.apache.org/jira/browse/FLINK-10556 > >>>>> ------------------------------------------------------------------ > >>>>> Sender:Xuefu <[hidden email]> > >>>>> Sent at:2018 Oct 25 (Thu) 01:08 > >>>>> Recipient:Xuefu <[hidden email]>; Shuyi Chen < > >>>>> [hidden email]> > >>>>> Cc:yanghua1127 <[hidden email]>; Fabian Hueske < > [hidden email]>; > >>>>> dev <[hidden email]>; user <[hidden email]> > >>>>> Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem > >>>>> > >>>>> Hi all, > >>>>> > >>>>> To wrap up the discussion, I have attached a PDF describing the > > >>>>> proposal, which is also attached to FLINK-10556 [1]. Please feel free to > >>>>> watch that JIRA to track the progress. > >>>>> > >>>>> Please also let me know if you have additional comments or questions. > >>>>> > >>>>> Thanks, > >>>>> Xuefu > >>>>> > >>>>> [1] https://issues.apache.org/jira/browse/FLINK-10556 > >>>>> > >>>>> > >>>>> ------------------------------------------------------------------ > >>>>> Sender:Xuefu <[hidden email]> > >>>>> Sent at:2018 Oct 16 (Tue) 03:40 > >>>>> Recipient:Shuyi Chen <[hidden email]> > >>>>> Cc:yanghua1127 <[hidden email]>; Fabian Hueske < > [hidden email]>; > >>>>> dev <[hidden email]>; user <[hidden email]> > >>>>> Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem > >>>>> > >>>>> Hi Shuyi, > >>>>> > > >>>>> Thank you for your input. Yes, I agreed with a phased approach and like > > >>>>> to move forward fast. :) We did some work internally on DDL utilizing babel > >>>>> parser in Calcite. While babel makes Calcite's grammar extensible, at > >>>>> first impression it still seems too cumbersome for a project when too > > >>>>> much extensions are made. It's even challenging to find where the extension > > >>>>> is needed! It would be certainly better if Calcite can magically support > > >>>>> Hive QL by just turning on a flag, such as that for MYSQL_5. I can also > > >>>>> see that this could mean a lot of work on Calcite. Nevertheless, I will > > >>>>> bring up the discussion over there and to see what their community thinks. > >>>>> > >>>>> Would mind to share more info about the proposal on DDL that you > >>>>> mentioned? We can certainly collaborate on this. > >>>>> > >>>>> Thanks, > >>>>> Xuefu > >>>>> > >>>>> ------------------------------------------------------------------ > >>>>> Sender:Shuyi Chen <[hidden email]> > >>>>> Sent at:2018 Oct 14 (Sun) 08:30 > >>>>> Recipient:Xuefu <[hidden email]> > >>>>> Cc:yanghua1127 <[hidden email]>; Fabian Hueske < > [hidden email]>; > >>>>> dev <[hidden email]>; user <[hidden email]> > >>>>> Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem > >>>>> > >>>>> Welcome to the community and thanks for the great proposal, Xuefu! I > > >>>>> think the proposal can be divided into 2 stages: making Flink to support > > >>>>> Hive features, and make Hive to work with Flink. I agreed with Timo that on > > >>>>> starting with a smaller scope, so we can make progress faster. As for [6], > > >>>>> a proposal for DDL is already in progress, and will come after the unified > > >>>>> SQL connector API is done. For supporting Hive syntax, we might need to > >>>>> work with the Calcite community, and a recent effort called babel ( > >>>>> https://issues.apache.org/jira/browse/CALCITE-2280) in Calcite might > >>>>> help here. > >>>>> > >>>>> Thanks > >>>>> Shuyi > >>>>> > >>>>> On Wed, Oct 10, 2018 at 8:02 PM Zhang, Xuefu < > [hidden email]> > >>>>> wrote: > >>>>> Hi Fabian/Vno, > >>>>> > > >>>>> Thank you very much for your encouragement inquiry. Sorry that I didn't > > >>>>> see Fabian's email until I read Vino's response just now. (Somehow Fabian's > >>>>> went to the spam folder.) > >>>>> > > >>>>> My proposal contains long-term and short-terms goals. Nevertheless, the > >>>>> effort will focus on the following areas, including Fabian's list: > >>>>> > >>>>> 1. Hive metastore connectivity - This covers both read/write access, > > >>>>> which means Flink can make full use of Hive's metastore as its catalog (at > >>>>> least for the batch but can extend for streaming as well). > > >>>>> 2. Metadata compatibility - Objects (databases, tables, partitions, etc) > > >>>>> created by Hive can be understood by Flink and the reverse direction is > >>>>> true also. > >>>>> 3. Data compatibility - Similar to #2, data produced by Hive can be > >>>>> consumed by Flink and vise versa. > > >>>>> 4. Support Hive UDFs - For all Hive's native udfs, Flink either provides > >>>>> its own implementation or make Hive's implementation work in Flink. > >>>>> Further, for user created UDFs in Hive, Flink SQL should provide a > > >>>>> mechanism allowing user to import them into Flink without any code change > >>>>> required. > >>>>> 5. Data types - Flink SQL should support all data types that are > >>>>> available in Hive. > >>>>> 6. SQL Language - Flink SQL should support SQL standard (such as > > >>>>> SQL2003) with extension to support Hive's syntax and language features, > >>>>> around DDL, DML, and SELECT queries. > > >>>>> 7. SQL CLI - this is currently developing in Flink but more effort is > >>>>> needed. > > >>>>> 8. Server - provide a server that's compatible with Hive's HiverServer2 > > >>>>> in thrift APIs, such that HiveServer2 users can reuse their existing client > >>>>> (such as beeline) but connect to Flink's thrift server instead. > > >>>>> 9. JDBC/ODBC drivers - Flink may provide its own JDBC/ODBC drivers for > >>>>> other application to use to connect to its thrift server > >>>>> 10. Support other user's customizations in Hive, such as Hive Serdes, > >>>>> storage handlers, etc. > > >>>>> 11. Better task failure tolerance and task scheduling at Flink runtime. > >>>>> > >>>>> As you can see, achieving all those requires significant effort and > > >>>>> across all layers in Flink. However, a short-term goal could include only > > >>>>> core areas (such as 1, 2, 4, 5, 6, 7) or start at a smaller scope (such as > >>>>> #3, #6). > >>>>> > > >>>>> Please share your further thoughts. If we generally agree that this is > > >>>>> the right direction, I could come up with a formal proposal quickly and > >>>>> then we can follow up with broader discussions. > >>>>> > >>>>> Thanks, > >>>>> Xuefu > >>>>> > >>>>> > >>>>> > >>>>> ------------------------------------------------------------------ > >>>>> Sender:vino yang <[hidden email]> > >>>>> Sent at:2018 Oct 11 (Thu) 09:45 > >>>>> Recipient:Fabian Hueske <[hidden email]> > >>>>> Cc:dev <[hidden email]>; Xuefu <[hidden email] > >; user < > >>>>> [hidden email]> > >>>>> Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem > >>>>> > >>>>> Hi Xuefu, > >>>>> > > >>>>> Appreciate this proposal, and like Fabian, it would look better if you > >>>>> can give more details of the plan. > >>>>> > >>>>> Thanks, vino. > >>>>> > >>>>> Fabian Hueske <[hidden email]> 于2018年10月10日周三 下午5:27写道： > >>>>> Hi Xuefu, > >>>>> > > >>>>> Welcome to the Flink community and thanks for starting this discussion! > >>>>> Better Hive integration would be really great! > >>>>> Can you go into details of what you are proposing? I can think of a > >>>>> couple ways to improve Flink in that regard: > >>>>> > >>>>> * Support for Hive UDFs > >>>>> * Support for Hive metadata catalog > >>>>> * Support for HiveQL syntax > >>>>> * ??? > >>>>> > >>>>> Best, Fabian > >>>>> > >>>>> Am Di., 9. Okt. 2018 um 19:22 Uhr schrieb Zhang, Xuefu < > >>>>> [hidden email]>: > >>>>> Hi all, > >>>>> > >>>>> Along with the community's effort, inside Alibaba we have explored > > >>>>> Flink's potential as an execution engine not just for stream processing but > >>>>> also for batch processing. We are encouraged by our findings and have > > >>>>> initiated our effort to make Flink's SQL capabilities full-fledged. When > > >>>>> comparing what's available in Flink to the offerings from competitive data > > >>>>> processing engines, we identified a major gap in Flink: a well integration > > >>>>> with Hive ecosystem. This is crucial to the success of Flink SQL and batch > > >>>>> due to the well-established data ecosystem around Hive. Therefore, we have > > >>>>> done some initial work along this direction but there are still a lot of > >>>>> effort needed. > >>>>> > >>>>> We have two strategies in mind. The first one is to make Flink SQL > > >>>>> full-fledged and well-integrated with Hive ecosystem. This is a similar > > >>>>> approach to what Spark SQL adopted. The second strategy is to make Hive > > >>>>> itself work with Flink, similar to the proposal in [1]. Each approach bears > > >>>>> its pros and cons, but they don’t need to be mutually exclusive with each > >>>>> targeting at different users and use cases. We believe that both will > >>>>> promote a much greater adoption of Flink beyond stream processing. > >>>>> > >>>>> We have been focused on the first approach and would like to showcase > > >>>>> Flink's batch and SQL capabilities with Flink SQL. However, we have also > >>>>> planned to start strategy #2 as the follow-up effort. > >>>>> > > >>>>> I'm completely new to Flink(, with a short bio [2] below), though many > > >>>>> of my colleagues here at Alibaba are long-time contributors. Nevertheless, > > >>>>> I'd like to share our thoughts and invite your early feedback. At the same > > >>>>> time, I am working on a detailed proposal on Flink SQL's integration with > >>>>> Hive ecosystem, which will be also shared when ready. > >>>>> > >>>>> While the ideas are simple, each approach will demand significant > > >>>>> effort, more than what we can afford. Thus, the input and contributions > >>>>> from the communities are greatly welcome and appreciated. > >>>>> > >>>>> Regards, > >>>>> > >>>>> > >>>>> Xuefu > >>>>> > >>>>> References: > >>>>> > >>>>> [1] https://issues.apache.org/jira/browse/HIVE-10712 > > >>>>> [2] Xuefu Zhang is a long-time open source veteran, worked or working on > >>>>> many projects under Apache Foundation, of which he is also an honored > > >>>>> member. About 10 years ago he worked in the Hadoop team at Yahoo where the > > >>>>> projects just got started. Later he worked at Cloudera, initiating and > > >>>>> leading the development of Hive on Spark project in the communities and > > >>>>> across many organizations. Prior to joining Alibaba, he worked at Uber > > >>>>> where he promoted Hive on Spark to all Uber's SQL on Hadoop workload and > >>>>> significantly improved Uber's cluster efficiency. > >>>>> > >>>>> > >>>>> > >>>>> > >>>>> -- > > >>>>> "So you have to trust that the dots will somehow connect in your future." > >>>>> > >>>>> > >>>>> -- > > >>>>> "So you have to trust that the dots will somehow connect in your future." > >>>>> > > |
Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem
|
|
Thanks, Bowen, for catching the error. I have granted comment permission with the link.
I also updated the doc with the latest class definitions. Everyone is encouraged to review and comment. Thanks, Xuefu ------------------------------------------------------------------ Sender:Bowen Li <[hidden email]> Sent at:2018 Nov 14 (Wed) 06:44 Recipient:Xuefu <[hidden email]> Cc:piotr <[hidden email]>; dev <[hidden email]>; Shuyi Chen <[hidden email]> Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem Hi Xuefu, Currently the new design doc is on “view only" mode, and people cannot leave comments. Can you please change it to "can comment" or "can edit" mode? Thanks, Bowen On Mon, Nov 12, 2018 at 9:51 PM Zhang, Xuefu <[hidden email]> wrote: Hi Piotr I have extracted the API portion of the design and the google doc is here. Please review and provide your feedback. Thanks, Xuefu ------------------------------------------------------------------ Sender:Xuefu <[hidden email]> Sent at:2018 Nov 12 (Mon) 12:43 Recipient:Piotr Nowojski <[hidden email]>; dev <[hidden email]> Cc:Bowen Li <[hidden email]>; Shuyi Chen <[hidden email]> Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem Hi Piotr, That sounds good to me. Let's close all the open questions ((there are a couple of them)) in the Google doc and I should be able to quickly split it into the three proposals as you suggested. Thanks, Xuefu ------------------------------------------------------------------ Sender:Piotr Nowojski <[hidden email]> Sent at:2018 Nov 9 (Fri) 22:46 Recipient:dev <[hidden email]>; Xuefu <[hidden email]> Cc:Bowen Li <[hidden email]>; Shuyi Chen <[hidden email]> Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem Hi, Yes, it seems like the best solution. Maybe someone else can also suggests if we can split it further? Maybe changes in the interface in one doc, reading from hive meta store another and final storing our meta informations in hive meta store? Piotrek > On 9 Nov 2018, at 01:44, Zhang, Xuefu <[hidden email]> wrote: > > Hi Piotr, > > That seems to be good idea! > > Since the google doc for the design is currently under extensive review, I will leave it as it is for now. However, I'll convert it to two different FLIPs when the time comes. > > How does it sound to you? > > Thanks, > Xuefu > > > ------------------------------------------------------------------ > Sender:Piotr Nowojski <[hidden email]> > Sent at:2018 Nov 9 (Fri) 02:31 > Recipient:dev <[hidden email]> > Cc:Bowen Li <[hidden email]>; Xuefu <[hidden email]>; Shuyi Chen <[hidden email]> > Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem > > Hi, > > Maybe we should split this topic (and the design doc) into couple of smaller ones, hopefully independent. The questions that you have asked Fabian have for example very little to do with reading metadata from Hive Meta Store? > > Piotrek > >> On 7 Nov 2018, at 14:27, Fabian Hueske <[hidden email]> wrote: >> >> Hi Xuefu and all, >> >> Thanks for sharing this design document! >> I'm very much in favor of restructuring / reworking the catalog handling in >> Flink SQL as outlined in the document. >> Most changes described in the design document seem to be rather general and >> not specifically related to the Hive integration. >> >> IMO, there are some aspects, especially those at the boundary of Hive and >> Flink, that need a bit more discussion. For example >> >> * What does it take to make Flink schema compatible with Hive schema? >> * How will Flink tables (descriptors) be stored in HMS? >> * How do both Hive catalogs differ? Could they be integrated into to a >> single one? When to use which one? >> * What meta information is provided by HMS? What of this can be leveraged >> by Flink? >> >> Thank you, >> Fabian >> >> Am Fr., 2. Nov. 2018 um 00:31 Uhr schrieb Bowen Li <[hidden email]>: >> >>> After taking a look at how other discussion threads work, I think it's >>> actually fine just keep our discussion here. It's up to you, Xuefu. >>> >>> The google doc LGTM. I left some minor comments. >>> >>> On Thu, Nov 1, 2018 at 10:17 AM Bowen Li <[hidden email]> wrote: >>> >>>> Hi all, >>>> >>>> As Xuefu has published the design doc on google, I agree with Shuyi's >>>> suggestion that we probably should start a new email thread like "[DISCUSS] >>>> ... Hive integration design ..." on only dev mailing list for community >>>> devs to review. The current thread sends to both dev and user list. >>>> >>>> This email thread is more like validating the general idea and direction >>>> with the community, and it's been pretty long and crowded so far. Since >>>> everyone is pro for the idea, we can move forward with another thread to >>>> discuss and finalize the design. >>>> >>>> Thanks, >>>> Bowen >>>> >>>> On Wed, Oct 31, 2018 at 12:16 PM Zhang, Xuefu <[hidden email]> >>>> wrote: >>>> >>>>> Hi Shuiyi, >>>>> >>>>> Good idea. Actually the PDF was converted from a google doc. Here is its >>>>> link: >>>>> >>>>> https://docs.google.com/document/d/1SkppRD_rE3uOKSN-LuZCqn4f7dz0zW5aa6T_hBZq5_o/edit?usp=sharing >>>>> Once we reach an agreement, I can convert it to a FLIP. >>>>> >>>>> Thanks, >>>>> Xuefu >>>>> >>>>> >>>>> >>>>> ------------------------------------------------------------------ >>>>> Sender:Shuyi Chen <[hidden email]> >>>>> Sent at:2018 Nov 1 (Thu) 02:47 >>>>> Recipient:Xuefu <[hidden email]> >>>>> Cc:vino yang <[hidden email]>; Fabian Hueske <[hidden email]>; >>>>> dev <[hidden email]>; user <[hidden email]> >>>>> Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem >>>>> >>>>> Hi Xuefu, >>>>> >>>>> Thanks a lot for driving this big effort. I would suggest convert your >>>>> proposal and design doc into a google doc, and share it on the dev mailing >>>>> list for the community to review and comment with title like "[DISCUSS] ... >>>>> Hive integration design ..." . Once approved, we can document it as a FLIP >>>>> (Flink Improvement Proposals), and use JIRAs to track the implementations. >>>>> What do you think? >>>>> >>>>> Shuyi >>>>> >>>>> On Tue, Oct 30, 2018 at 11:32 AM Zhang, Xuefu <[hidden email]> >>>>> wrote: >>>>> Hi all, >>>>> >>>>> I have also shared a design doc on Hive metastore integration that is >>>>> attached here and also to FLINK-10556[1]. Please kindly review and share >>>>> your feedback. >>>>> >>>>> >>>>> Thanks, >>>>> Xuefu >>>>> >>>>> [1] https://issues.apache.org/jira/browse/FLINK-10556 >>>>> ------------------------------------------------------------------ >>>>> Sender:Xuefu <[hidden email]> >>>>> Sent at:2018 Oct 25 (Thu) 01:08 >>>>> Recipient:Xuefu <[hidden email]>; Shuyi Chen < >>>>> [hidden email]> >>>>> Cc:yanghua1127 <[hidden email]>; Fabian Hueske <[hidden email]>; >>>>> dev <[hidden email]>; user <[hidden email]> >>>>> Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem >>>>> >>>>> Hi all, >>>>> >>>>> To wrap up the discussion, I have attached a PDF describing the >>>>> proposal, which is also attached to FLINK-10556 [1]. Please feel free to >>>>> watch that JIRA to track the progress. >>>>> >>>>> Please also let me know if you have additional comments or questions. >>>>> >>>>> Thanks, >>>>> Xuefu >>>>> >>>>> [1] https://issues.apache.org/jira/browse/FLINK-10556 >>>>> >>>>> >>>>> ------------------------------------------------------------------ >>>>> Sender:Xuefu <[hidden email]> >>>>> Sent at:2018 Oct 16 (Tue) 03:40 >>>>> Recipient:Shuyi Chen <[hidden email]> >>>>> Cc:yanghua1127 <[hidden email]>; Fabian Hueske <[hidden email]>; >>>>> dev <[hidden email]>; user <[hidden email]> >>>>> Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem >>>>> >>>>> Hi Shuyi, >>>>> >>>>> Thank you for your input. Yes, I agreed with a phased approach and like >>>>> to move forward fast. :) We did some work internally on DDL utilizing babel >>>>> parser in Calcite. While babel makes Calcite's grammar extensible, at >>>>> first impression it still seems too cumbersome for a project when too >>>>> much extensions are made. It's even challenging to find where the extension >>>>> is needed! It would be certainly better if Calcite can magically support >>>>> Hive QL by just turning on a flag, such as that for MYSQL_5. I can also >>>>> see that this could mean a lot of work on Calcite. Nevertheless, I will >>>>> bring up the discussion over there and to see what their community thinks. >>>>> >>>>> Would mind to share more info about the proposal on DDL that you >>>>> mentioned? We can certainly collaborate on this. >>>>> >>>>> Thanks, >>>>> Xuefu >>>>> >>>>> ------------------------------------------------------------------ >>>>> Sender:Shuyi Chen <[hidden email]> >>>>> Sent at:2018 Oct 14 (Sun) 08:30 >>>>> Recipient:Xuefu <[hidden email]> >>>>> Cc:yanghua1127 <[hidden email]>; Fabian Hueske <[hidden email]>; >>>>> dev <[hidden email]>; user <[hidden email]> >>>>> Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem >>>>> >>>>> Welcome to the community and thanks for the great proposal, Xuefu! I >>>>> think the proposal can be divided into 2 stages: making Flink to support >>>>> Hive features, and make Hive to work with Flink. I agreed with Timo that on >>>>> starting with a smaller scope, so we can make progress faster. As for [6], >>>>> a proposal for DDL is already in progress, and will come after the unified >>>>> SQL connector API is done. For supporting Hive syntax, we might need to >>>>> work with the Calcite community, and a recent effort called babel ( >>>>> https://issues.apache.org/jira/browse/CALCITE-2280) in Calcite might >>>>> help here. >>>>> >>>>> Thanks >>>>> Shuyi >>>>> >>>>> On Wed, Oct 10, 2018 at 8:02 PM Zhang, Xuefu <[hidden email]> >>>>> wrote: >>>>> Hi Fabian/Vno, >>>>> >>>>> Thank you very much for your encouragement inquiry. Sorry that I didn't >>>>> see Fabian's email until I read Vino's response just now. (Somehow Fabian's >>>>> went to the spam folder.) >>>>> >>>>> My proposal contains long-term and short-terms goals. Nevertheless, the >>>>> effort will focus on the following areas, including Fabian's list: >>>>> >>>>> 1. Hive metastore connectivity - This covers both read/write access, >>>>> which means Flink can make full use of Hive's metastore as its catalog (at >>>>> least for the batch but can extend for streaming as well). >>>>> 2. Metadata compatibility - Objects (databases, tables, partitions, etc) >>>>> created by Hive can be understood by Flink and the reverse direction is >>>>> true also. >>>>> 3. Data compatibility - Similar to #2, data produced by Hive can be >>>>> consumed by Flink and vise versa. >>>>> 4. Support Hive UDFs - For all Hive's native udfs, Flink either provides >>>>> its own implementation or make Hive's implementation work in Flink. >>>>> Further, for user created UDFs in Hive, Flink SQL should provide a >>>>> mechanism allowing user to import them into Flink without any code change >>>>> required. >>>>> 5. Data types - Flink SQL should support all data types that are >>>>> available in Hive. >>>>> 6. SQL Language - Flink SQL should support SQL standard (such as >>>>> SQL2003) with extension to support Hive's syntax and language features, >>>>> around DDL, DML, and SELECT queries. >>>>> 7. SQL CLI - this is currently developing in Flink but more effort is >>>>> needed. >>>>> 8. Server - provide a server that's compatible with Hive's HiverServer2 >>>>> in thrift APIs, such that HiveServer2 users can reuse their existing client >>>>> (such as beeline) but connect to Flink's thrift server instead. >>>>> 9. JDBC/ODBC drivers - Flink may provide its own JDBC/ODBC drivers for >>>>> other application to use to connect to its thrift server >>>>> 10. Support other user's customizations in Hive, such as Hive Serdes, >>>>> storage handlers, etc. >>>>> 11. Better task failure tolerance and task scheduling at Flink runtime. >>>>> >>>>> As you can see, achieving all those requires significant effort and >>>>> across all layers in Flink. However, a short-term goal could include only >>>>> core areas (such as 1, 2, 4, 5, 6, 7) or start at a smaller scope (such as >>>>> #3, #6). >>>>> >>>>> Please share your further thoughts. If we generally agree that this is >>>>> the right direction, I could come up with a formal proposal quickly and >>>>> then we can follow up with broader discussions. >>>>> >>>>> Thanks, >>>>> Xuefu >>>>> >>>>> >>>>> >>>>> ------------------------------------------------------------------ >>>>> Sender:vino yang <[hidden email]> >>>>> Sent at:2018 Oct 11 (Thu) 09:45 >>>>> Recipient:Fabian Hueske <[hidden email]> >>>>> Cc:dev <[hidden email]>; Xuefu <[hidden email]>; user < >>>>> [hidden email]> >>>>> Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem >>>>> >>>>> Hi Xuefu, >>>>> >>>>> Appreciate this proposal, and like Fabian, it would look better if you >>>>> can give more details of the plan. >>>>> >>>>> Thanks, vino. >>>>> >>>>> Fabian Hueske <[hidden email]> 于2018年10月10日周三 下午5:27写道： >>>>> Hi Xuefu, >>>>> >>>>> Welcome to the Flink community and thanks for starting this discussion! >>>>> Better Hive integration would be really great! >>>>> Can you go into details of what you are proposing? I can think of a >>>>> couple ways to improve Flink in that regard: >>>>> >>>>> * Support for Hive UDFs >>>>> * Support for Hive metadata catalog >>>>> * Support for HiveQL syntax >>>>> * ??? >>>>> >>>>> Best, Fabian >>>>> >>>>> Am Di., 9. Okt. 2018 um 19:22 Uhr schrieb Zhang, Xuefu < >>>>> [hidden email]>: >>>>> Hi all, >>>>> >>>>> Along with the community's effort, inside Alibaba we have explored >>>>> Flink's potential as an execution engine not just for stream processing but >>>>> also for batch processing. We are encouraged by our findings and have >>>>> initiated our effort to make Flink's SQL capabilities full-fledged. When >>>>> comparing what's available in Flink to the offerings from competitive data >>>>> processing engines, we identified a major gap in Flink: a well integration >>>>> with Hive ecosystem. This is crucial to the success of Flink SQL and batch >>>>> due to the well-established data ecosystem around Hive. Therefore, we have >>>>> done some initial work along this direction but there are still a lot of >>>>> effort needed. >>>>> >>>>> We have two strategies in mind. The first one is to make Flink SQL >>>>> full-fledged and well-integrated with Hive ecosystem. This is a similar >>>>> approach to what Spark SQL adopted. The second strategy is to make Hive >>>>> itself work with Flink, similar to the proposal in [1]. Each approach bears >>>>> its pros and cons, but they don’t need to be mutually exclusive with each >>>>> targeting at different users and use cases. We believe that both will >>>>> promote a much greater adoption of Flink beyond stream processing. >>>>> >>>>> We have been focused on the first approach and would like to showcase >>>>> Flink's batch and SQL capabilities with Flink SQL. However, we have also >>>>> planned to start strategy #2 as the follow-up effort. >>>>> >>>>> I'm completely new to Flink(, with a short bio [2] below), though many >>>>> of my colleagues here at Alibaba are long-time contributors. Nevertheless, >>>>> I'd like to share our thoughts and invite your early feedback. At the same >>>>> time, I am working on a detailed proposal on Flink SQL's integration with >>>>> Hive ecosystem, which will be also shared when ready. >>>>> >>>>> While the ideas are simple, each approach will demand significant >>>>> effort, more than what we can afford. Thus, the input and contributions >>>>> from the communities are greatly welcome and appreciated. >>>>> >>>>> Regards, >>>>> >>>>> >>>>> Xuefu >>>>> >>>>> References: >>>>> >>>>> [1] https://issues.apache.org/jira/browse/HIVE-10712 >>>>> [2] Xuefu Zhang is a long-time open source veteran, worked or working on >>>>> many projects under Apache Foundation, of which he is also an honored >>>>> member. About 10 years ago he worked in the Hadoop team at Yahoo where the >>>>> projects just got started. Later he worked at Cloudera, initiating and >>>>> leading the development of Hive on Spark project in the communities and >>>>> across many organizations. Prior to joining Alibaba, he worked at Uber >>>>> where he promoted Hive on Spark to all Uber's SQL on Hadoop workload and >>>>> significantly improved Uber's cluster efficiency. >>>>> >>>>> >>>>> >>>>> >>>>> -- >>>>> "So you have to trust that the dots will somehow connect in your future." >>>>> >>>>> >>>>> -- >>>>> "So you have to trust that the dots will somehow connect in your future." >>>>> |
|
|
Thanks for keeping on improving the overall design, Xuefu! It looks quite
good to me now. Would be nice that cc-ed Flink committers can help to review and confirm! One minor suggestion: Since the last section of design doc already touches some new sql statements, shall we add another section in our doc and formalize the new sql statements in SQL Client and TableEnvironment that are gonna come along naturally with our design? Here are some that the design doc mentioned and some that I came up with: To be added: - USE <catalog> - set default catalog - USE <catalog.schema> - set default schema - SHOW CATALOGS - show all registered catalogs - SHOW SCHEMAS [FROM catalog] - list schemas in the current default catalog or the specified catalog - DESCRIBE VIEW view - show the view's definition in CatalogView - SHOW VIEWS [FROM schema/catalog.schema] - show views from current or a specified schema. (DDLs that can be addressed by either our design or Shuyi's DDL design) - CREATE/DROP/ALTER SCHEMA schema - CREATE/DROP/ALTER CATALOG catalog To be modified: - SHOW TABLES [FROM schema/catalog.schema] - show tables from current or a specified schema. Add 'from schema' to existing 'SHOW TABLES' statement - SHOW FUNCTIONS [FROM schema/catalog.schema] - show functions from current or a specified schema. Add 'from schema' to existing 'SHOW TABLES' statement' Thanks, Bowen On Wed, Nov 14, 2018 at 10:39 PM Zhang, Xuefu <[hidden email]> wrote: > Thanks, Bowen, for catching the error. I have granted comment permission > with the link. > > I also updated the doc with the latest class definitions. Everyone is > encouraged to review and comment. > > Thanks, > Xuefu > > ------------------------------------------------------------------ > Sender:Bowen Li <[hidden email]> > Sent at:2018 Nov 14 (Wed) 06:44 > Recipient:Xuefu <[hidden email]> > Cc:piotr <[hidden email]>; dev <[hidden email]>; Shuyi > Chen <[hidden email]> > Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem > > Hi Xuefu, > > Currently the new design doc > <https://docs.google.com/document/d/1Y9it78yaUvbv4g572ZK_lZnZaAGjqwM_EhjdOv4yJtw/edit> > is on “view only" mode, and people cannot leave comments. Can you please > change it to "can comment" or "can edit" mode? > > Thanks, Bowen > > > On Mon, Nov 12, 2018 at 9:51 PM Zhang, Xuefu <[hidden email]> > wrote: > Hi Piotr > > I have extracted the API portion of the design and the google doc is here > <https://docs.google.com/document/d/1Y9it78yaUvbv4g572ZK_lZnZaAGjqwM_EhjdOv4yJtw/edit?usp=sharing>. > Please review and provide your feedback. > > Thanks, > Xuefu > > ------------------------------------------------------------------ > Sender:Xuefu <[hidden email]> > Sent at:2018 Nov 12 (Mon) 12:43 > Recipient:Piotr Nowojski <[hidden email]>; dev < > [hidden email]> > Cc:Bowen Li <[hidden email]>; Shuyi Chen <[hidden email]> > Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem > > Hi Piotr, > > That sounds good to me. Let's close all the open questions ((there are a > couple of them)) in the Google doc and I should be able to quickly split > it into the three proposals as you suggested. > > Thanks, > Xuefu > > ------------------------------------------------------------------ > Sender:Piotr Nowojski <[hidden email]> > Sent at:2018 Nov 9 (Fri) 22:46 > Recipient:dev <[hidden email]>; Xuefu <[hidden email]> > Cc:Bowen Li <[hidden email]>; Shuyi Chen <[hidden email]> > Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem > > Hi, > > > Yes, it seems like the best solution. Maybe someone else can also suggests if we can split it further? Maybe changes in the interface in one doc, reading from hive meta store another and final storing our meta informations in hive meta store? > > Piotrek > > > On 9 Nov 2018, at 01:44, Zhang, Xuefu <[hidden email]> wrote: > > > > Hi Piotr, > > > > That seems to be good idea! > > > > > Since the google doc for the design is currently under extensive review, I will leave it as it is for now. However, I'll convert it to two different FLIPs when the time comes. > > > > How does it sound to you? > > > > Thanks, > > Xuefu > > > > > > ------------------------------------------------------------------ > > Sender:Piotr Nowojski <[hidden email]> > > Sent at:2018 Nov 9 (Fri) 02:31 > > Recipient:dev <[hidden email]> > > Cc:Bowen Li <[hidden email]>; Xuefu <[hidden email] > >; Shuyi Chen <[hidden email]> > > Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem > > > > Hi, > > > > > Maybe we should split this topic (and the design doc) into couple of smaller ones, hopefully independent. The questions that you have asked Fabian have for example very little to do with reading metadata from Hive Meta Store? > > > > Piotrek > > > >> On 7 Nov 2018, at 14:27, Fabian Hueske <[hidden email]> wrote: > >> > >> Hi Xuefu and all, > >> > >> Thanks for sharing this design document! > > >> I'm very much in favor of restructuring / reworking the catalog handling in > >> Flink SQL as outlined in the document. > > >> Most changes described in the design document seem to be rather general and > >> not specifically related to the Hive integration. > >> > > >> IMO, there are some aspects, especially those at the boundary of Hive and > >> Flink, that need a bit more discussion. For example > >> > >> * What does it take to make Flink schema compatible with Hive schema? > >> * How will Flink tables (descriptors) be stored in HMS? > >> * How do both Hive catalogs differ? Could they be integrated into to a > >> single one? When to use which one? > > >> * What meta information is provided by HMS? What of this can be leveraged > >> by Flink? > >> > >> Thank you, > >> Fabian > >> > >> Am Fr., 2. Nov. 2018 um 00:31 Uhr schrieb Bowen Li <[hidden email] > >: > >> > >>> After taking a look at how other discussion threads work, I think it's > >>> actually fine just keep our discussion here. It's up to you, Xuefu. > >>> > >>> The google doc LGTM. I left some minor comments. > >>> > >>> On Thu, Nov 1, 2018 at 10:17 AM Bowen Li <[hidden email]> wrote: > >>> > >>>> Hi all, > >>>> > >>>> As Xuefu has published the design doc on google, I agree with Shuyi's > > >>>> suggestion that we probably should start a new email thread like "[DISCUSS] > > >>>> ... Hive integration design ..." on only dev mailing list for community > >>>> devs to review. The current thread sends to both dev and user list. > >>>> > > >>>> This email thread is more like validating the general idea and direction > > >>>> with the community, and it's been pretty long and crowded so far. Since > > >>>> everyone is pro for the idea, we can move forward with another thread to > >>>> discuss and finalize the design. > >>>> > >>>> Thanks, > >>>> Bowen > >>>> > >>>> On Wed, Oct 31, 2018 at 12:16 PM Zhang, Xuefu < > [hidden email]> > >>>> wrote: > >>>> > >>>>> Hi Shuiyi, > >>>>> > > >>>>> Good idea. Actually the PDF was converted from a google doc. Here is its > >>>>> link: > >>>>> > >>>>> > https://docs.google.com/document/d/1SkppRD_rE3uOKSN-LuZCqn4f7dz0zW5aa6T_hBZq5_o/edit?usp=sharing > >>>>> Once we reach an agreement, I can convert it to a FLIP. > >>>>> > >>>>> Thanks, > >>>>> Xuefu > >>>>> > >>>>> > >>>>> > >>>>> ------------------------------------------------------------------ > >>>>> Sender:Shuyi Chen <[hidden email]> > >>>>> Sent at:2018 Nov 1 (Thu) 02:47 > >>>>> Recipient:Xuefu <[hidden email]> > >>>>> Cc:vino yang <[hidden email]>; Fabian Hueske < > [hidden email]>; > >>>>> dev <[hidden email]>; user <[hidden email]> > >>>>> Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem > >>>>> > >>>>> Hi Xuefu, > >>>>> > > >>>>> Thanks a lot for driving this big effort. I would suggest convert your > > >>>>> proposal and design doc into a google doc, and share it on the dev mailing > > >>>>> list for the community to review and comment with title like "[DISCUSS] ... > > >>>>> Hive integration design ..." . Once approved, we can document it as a FLIP > > >>>>> (Flink Improvement Proposals), and use JIRAs to track the implementations. > >>>>> What do you think? > >>>>> > >>>>> Shuyi > >>>>> > >>>>> On Tue, Oct 30, 2018 at 11:32 AM Zhang, Xuefu < > [hidden email]> > >>>>> wrote: > >>>>> Hi all, > >>>>> > >>>>> I have also shared a design doc on Hive metastore integration that is > > >>>>> attached here and also to FLINK-10556[1]. Please kindly review and share > >>>>> your feedback. > >>>>> > >>>>> > >>>>> Thanks, > >>>>> Xuefu > >>>>> > >>>>> [1] https://issues.apache.org/jira/browse/FLINK-10556 > >>>>> ------------------------------------------------------------------ > >>>>> Sender:Xuefu <[hidden email]> > >>>>> Sent at:2018 Oct 25 (Thu) 01:08 > >>>>> Recipient:Xuefu <[hidden email]>; Shuyi Chen < > >>>>> [hidden email]> > >>>>> Cc:yanghua1127 <[hidden email]>; Fabian Hueske < > [hidden email]>; > >>>>> dev <[hidden email]>; user <[hidden email]> > >>>>> Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem > >>>>> > >>>>> Hi all, > >>>>> > >>>>> To wrap up the discussion, I have attached a PDF describing the > > >>>>> proposal, which is also attached to FLINK-10556 [1]. Please feel free to > >>>>> watch that JIRA to track the progress. > >>>>> > >>>>> Please also let me know if you have additional comments or questions. > >>>>> > >>>>> Thanks, > >>>>> Xuefu > >>>>> > >>>>> [1] https://issues.apache.org/jira/browse/FLINK-10556 > >>>>> > >>>>> > >>>>> ------------------------------------------------------------------ > >>>>> Sender:Xuefu <[hidden email]> > >>>>> Sent at:2018 Oct 16 (Tue) 03:40 > >>>>> Recipient:Shuyi Chen <[hidden email]> > >>>>> Cc:yanghua1127 <[hidden email]>; Fabian Hueske < > [hidden email]>; > >>>>> dev <[hidden email]>; user <[hidden email]> > >>>>> Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem > >>>>> > >>>>> Hi Shuyi, > >>>>> > > >>>>> Thank you for your input. Yes, I agreed with a phased approach and like > > >>>>> to move forward fast. :) We did some work internally on DDL utilizing babel > >>>>> parser in Calcite. While babel makes Calcite's grammar extensible, at > >>>>> first impression it still seems too cumbersome for a project when too > > >>>>> much extensions are made. It's even challenging to find where the extension > > >>>>> is needed! It would be certainly better if Calcite can magically support > > >>>>> Hive QL by just turning on a flag, such as that for MYSQL_5. I can also > > >>>>> see that this could mean a lot of work on Calcite. Nevertheless, I will > > >>>>> bring up the discussion over there and to see what their community thinks. > >>>>> > >>>>> Would mind to share more info about the proposal on DDL that you > >>>>> mentioned? We can certainly collaborate on this. > >>>>> > >>>>> Thanks, > >>>>> Xuefu > >>>>> > >>>>> ------------------------------------------------------------------ > >>>>> Sender:Shuyi Chen <[hidden email]> > >>>>> Sent at:2018 Oct 14 (Sun) 08:30 > >>>>> Recipient:Xuefu <[hidden email]> > >>>>> Cc:yanghua1127 <[hidden email]>; Fabian Hueske < > [hidden email]>; > >>>>> dev <[hidden email]>; user <[hidden email]> > >>>>> Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem > >>>>> > >>>>> Welcome to the community and thanks for the great proposal, Xuefu! I > > >>>>> think the proposal can be divided into 2 stages: making Flink to support > > >>>>> Hive features, and make Hive to work with Flink. I agreed with Timo that on > > >>>>> starting with a smaller scope, so we can make progress faster. As for [6], > > >>>>> a proposal for DDL is already in progress, and will come after the unified > > >>>>> SQL connector API is done. For supporting Hive syntax, we might need to > >>>>> work with the Calcite community, and a recent effort called babel ( > >>>>> https://issues.apache.org/jira/browse/CALCITE-2280) in Calcite might > >>>>> help here. > >>>>> > >>>>> Thanks > >>>>> Shuyi > >>>>> > >>>>> On Wed, Oct 10, 2018 at 8:02 PM Zhang, Xuefu < > [hidden email]> > >>>>> wrote: > >>>>> Hi Fabian/Vno, > >>>>> > > >>>>> Thank you very much for your encouragement inquiry. Sorry that I didn't > > >>>>> see Fabian's email until I read Vino's response just now. (Somehow Fabian's > >>>>> went to the spam folder.) > >>>>> > > >>>>> My proposal contains long-term and short-terms goals. Nevertheless, the > >>>>> effort will focus on the following areas, including Fabian's list: > >>>>> > >>>>> 1. Hive metastore connectivity - This covers both read/write access, > > >>>>> which means Flink can make full use of Hive's metastore as its catalog (at > >>>>> least for the batch but can extend for streaming as well). > > >>>>> 2. Metadata compatibility - Objects (databases, tables, partitions, etc) > > >>>>> created by Hive can be understood by Flink and the reverse direction is > >>>>> true also. > >>>>> 3. Data compatibility - Similar to #2, data produced by Hive can be > >>>>> consumed by Flink and vise versa. > > >>>>> 4. Support Hive UDFs - For all Hive's native udfs, Flink either provides > >>>>> its own implementation or make Hive's implementation work in Flink. > >>>>> Further, for user created UDFs in Hive, Flink SQL should provide a > > >>>>> mechanism allowing user to import them into Flink without any code change > >>>>> required. > >>>>> 5. Data types - Flink SQL should support all data types that are > >>>>> available in Hive. > >>>>> 6. SQL Language - Flink SQL should support SQL standard (such as > > >>>>> SQL2003) with extension to support Hive's syntax and language features, > >>>>> around DDL, DML, and SELECT queries. > > >>>>> 7. SQL CLI - this is currently developing in Flink but more effort is > >>>>> needed. > > >>>>> 8. Server - provide a server that's compatible with Hive's HiverServer2 > > >>>>> in thrift APIs, such that HiveServer2 users can reuse their existing client > >>>>> (such as beeline) but connect to Flink's thrift server instead. > > >>>>> 9. JDBC/ODBC drivers - Flink may provide its own JDBC/ODBC drivers for > >>>>> other application to use to connect to its thrift server > >>>>> 10. Support other user's customizations in Hive, such as Hive Serdes, > >>>>> storage handlers, etc. > > >>>>> 11. Better task failure tolerance and task scheduling at Flink runtime. > >>>>> > >>>>> As you can see, achieving all those requires significant effort and > > >>>>> across all layers in Flink. However, a short-term goal could include only > > >>>>> core areas (such as 1, 2, 4, 5, 6, 7) or start at a smaller scope (such as > >>>>> #3, #6). > >>>>> > > >>>>> Please share your further thoughts. If we generally agree that this is > > >>>>> the right direction, I could come up with a formal proposal quickly and > >>>>> then we can follow up with broader discussions. > >>>>> > >>>>> Thanks, > >>>>> Xuefu > >>>>> > >>>>> > >>>>> > >>>>> ------------------------------------------------------------------ > >>>>> Sender:vino yang <[hidden email]> > >>>>> Sent at:2018 Oct 11 (Thu) 09:45 > >>>>> Recipient:Fabian Hueske <[hidden email]> > >>>>> Cc:dev <[hidden email]>; Xuefu <[hidden email] > >; user < > >>>>> [hidden email]> > >>>>> Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem > >>>>> > >>>>> Hi Xuefu, > >>>>> > > >>>>> Appreciate this proposal, and like Fabian, it would look better if you > >>>>> can give more details of the plan. > >>>>> > >>>>> Thanks, vino. > >>>>> > >>>>> Fabian Hueske <[hidden email]> 于2018年10月10日周三 下午5:27写道： > >>>>> Hi Xuefu, > >>>>> > > >>>>> Welcome to the Flink community and thanks for starting this discussion! > >>>>> Better Hive integration would be really great! > >>>>> Can you go into details of what you are proposing? I can think of a > >>>>> couple ways to improve Flink in that regard: > >>>>> > >>>>> * Support for Hive UDFs > >>>>> * Support for Hive metadata catalog > >>>>> * Support for HiveQL syntax > >>>>> * ??? > >>>>> > >>>>> Best, Fabian > >>>>> > >>>>> Am Di., 9. Okt. 2018 um 19:22 Uhr schrieb Zhang, Xuefu < > >>>>> [hidden email]>: > >>>>> Hi all, > >>>>> > >>>>> Along with the community's effort, inside Alibaba we have explored > > >>>>> Flink's potential as an execution engine not just for stream processing but > >>>>> also for batch processing. We are encouraged by our findings and have > > >>>>> initiated our effort to make Flink's SQL capabilities full-fledged. When > > >>>>> comparing what's available in Flink to the offerings from competitive data > > >>>>> processing engines, we identified a major gap in Flink: a well integration > > >>>>> with Hive ecosystem. This is crucial to the success of Flink SQL and batch > > >>>>> due to the well-established data ecosystem around Hive. Therefore, we have > > >>>>> done some initial work along this direction but there are still a lot of > >>>>> effort needed. > >>>>> > >>>>> We have two strategies in mind. The first one is to make Flink SQL > > >>>>> full-fledged and well-integrated with Hive ecosystem. This is a similar > > >>>>> approach to what Spark SQL adopted. The second strategy is to make Hive > > >>>>> itself work with Flink, similar to the proposal in [1]. Each approach bears > > >>>>> its pros and cons, but they don’t need to be mutually exclusive with each > >>>>> targeting at different users and use cases. We believe that both will > >>>>> promote a much greater adoption of Flink beyond stream processing. > >>>>> > >>>>> We have been focused on the first approach and would like to showcase > > >>>>> Flink's batch and SQL capabilities with Flink SQL. However, we have also > >>>>> planned to start strategy #2 as the follow-up effort. > >>>>> > > >>>>> I'm completely new to Flink(, with a short bio [2] below), though many > > >>>>> of my colleagues here at Alibaba are long-time contributors. Nevertheless, > > >>>>> I'd like to share our thoughts and invite your early feedback. At the same > > >>>>> time, I am working on a detailed proposal on Flink SQL's integration with > >>>>> Hive ecosystem, which will be also shared when ready. > >>>>> > >>>>> While the ideas are simple, each approach will demand significant > > >>>>> effort, more than what we can afford. Thus, the input and contributions > >>>>> from the communities are greatly welcome and appreciated. > >>>>> > >>>>> Regards, > >>>>> > >>>>> > >>>>> Xuefu > >>>>> > >>>>> References: > >>>>> > >>>>> [1] https://issues.apache.org/jira/browse/HIVE-10712 > > >>>>> [2] Xuefu Zhang is a long-time open source veteran, worked or working on > >>>>> many projects under Apache Foundation, of which he is also an honored > > >>>>> member. About 10 years ago he worked in the Hadoop team at Yahoo where the > > >>>>> projects just got started. Later he worked at Cloudera, initiating and > > >>>>> leading the development of Hive on Spark project in the communities and > > >>>>> across many organizations. Prior to joining Alibaba, he worked at Uber > > >>>>> where he promoted Hive on Spark to all Uber's SQL on Hadoop workload and > >>>>> significantly improved Uber's cluster efficiency. > >>>>> > >>>>> > >>>>> > >>>>> > >>>>> -- > > >>>>> "So you have to trust that the dots will somehow connect in your future." > >>>>> > >>>>> > >>>>> -- > > >>>>> "So you have to trust that the dots will somehow connect in your future." > >>>>> > > |
Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem
|
|
Thanks Xuefu for the detailed design doc! One question on the properties
associated with the catalog objects. Are we going to leave them completely free form or we are going to set some standard for that? I think that the answer may depend on if we want to explore catalog specific optimization opportunities. In any case, I think that it might be helpful for standardize as much as possible into strongly typed classes and use leave these properties for catalog specific things. But I think that we can do it in steps. Xiaowei On Fri, Nov 16, 2018 at 4:00 AM Bowen Li <[hidden email]> wrote: > Thanks for keeping on improving the overall design, Xuefu! It looks quite > good to me now. > > Would be nice that cc-ed Flink committers can help to review and confirm! > > > > One minor suggestion: Since the last section of design doc already touches > some new sql statements, shall we add another section in our doc and > formalize the new sql statements in SQL Client and TableEnvironment that > are gonna come along naturally with our design? Here are some that the > design doc mentioned and some that I came up with: > > To be added: > > - USE <catalog> - set default catalog > - USE <catalog.schema> - set default schema > - SHOW CATALOGS - show all registered catalogs > - SHOW SCHEMAS [FROM catalog] - list schemas in the current default > catalog or the specified catalog > - DESCRIBE VIEW view - show the view's definition in CatalogView > - SHOW VIEWS [FROM schema/catalog.schema] - show views from current or a > specified schema. > > (DDLs that can be addressed by either our design or Shuyi's DDL design) > > - CREATE/DROP/ALTER SCHEMA schema > - CREATE/DROP/ALTER CATALOG catalog > > To be modified: > > - SHOW TABLES [FROM schema/catalog.schema] - show tables from current or > a specified schema. Add 'from schema' to existing 'SHOW TABLES' > statement > - SHOW FUNCTIONS [FROM schema/catalog.schema] - show functions from > current or a specified schema. Add 'from schema' to existing 'SHOW > TABLES' > statement' > > > Thanks, Bowen > > > > On Wed, Nov 14, 2018 at 10:39 PM Zhang, Xuefu <[hidden email]> > wrote: > > > Thanks, Bowen, for catching the error. I have granted comment permission > > with the link. > > > > I also updated the doc with the latest class definitions. Everyone is > > encouraged to review and comment. > > > > Thanks, > > Xuefu > > > > ------------------------------------------------------------------ > > Sender:Bowen Li <[hidden email]> > > Sent at:2018 Nov 14 (Wed) 06:44 > > Recipient:Xuefu <[hidden email]> > > Cc:piotr <[hidden email]>; dev <[hidden email]>; Shuyi > > Chen <[hidden email]> > > Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem > > > > Hi Xuefu, > > > > Currently the new design doc > > < > https://docs.google.com/document/d/1Y9it78yaUvbv4g572ZK_lZnZaAGjqwM_EhjdOv4yJtw/edit > > > > is on “view only" mode, and people cannot leave comments. Can you please > > change it to "can comment" or "can edit" mode? > > > > Thanks, Bowen > > > > > > On Mon, Nov 12, 2018 at 9:51 PM Zhang, Xuefu <[hidden email]> > > wrote: > > Hi Piotr > > > > I have extracted the API portion of the design and the google doc is > here > > < > https://docs.google.com/document/d/1Y9it78yaUvbv4g572ZK_lZnZaAGjqwM_EhjdOv4yJtw/edit?usp=sharing > >. > > Please review and provide your feedback. > > > > Thanks, > > Xuefu > > > > ------------------------------------------------------------------ > > Sender:Xuefu <[hidden email]> > > Sent at:2018 Nov 12 (Mon) 12:43 > > Recipient:Piotr Nowojski <[hidden email]>; dev < > > [hidden email]> > > Cc:Bowen Li <[hidden email]>; Shuyi Chen <[hidden email]> > > Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem > > > > Hi Piotr, > > > > That sounds good to me. Let's close all the open questions ((there are a > > couple of them)) in the Google doc and I should be able to quickly split > > it into the three proposals as you suggested. > > > > Thanks, > > Xuefu > > > > ------------------------------------------------------------------ > > Sender:Piotr Nowojski <[hidden email]> > > Sent at:2018 Nov 9 (Fri) 22:46 > > Recipient:dev <[hidden email]>; Xuefu <[hidden email]> > > Cc:Bowen Li <[hidden email]>; Shuyi Chen <[hidden email]> > > Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem > > > > Hi, > > > > > > Yes, it seems like the best solution. Maybe someone else can also > suggests if we can split it further? Maybe changes in the interface in one > doc, reading from hive meta store another and final storing our meta > informations in hive meta store? > > > > Piotrek > > > > > On 9 Nov 2018, at 01:44, Zhang, Xuefu <[hidden email]> wrote: > > > > > > Hi Piotr, > > > > > > That seems to be good idea! > > > > > > > > Since the google doc for the design is currently under extensive > review, I will leave it as it is for now. However, I'll convert it to two > different FLIPs when the time comes. > > > > > > How does it sound to you? > > > > > > Thanks, > > > Xuefu > > > > > > > > > ------------------------------------------------------------------ > > > Sender:Piotr Nowojski <[hidden email]> > > > Sent at:2018 Nov 9 (Fri) 02:31 > > > Recipient:dev <[hidden email]> > > > Cc:Bowen Li <[hidden email]>; Xuefu <[hidden email] > > >; Shuyi Chen <[hidden email]> > > > Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem > > > > > > Hi, > > > > > > > > Maybe we should split this topic (and the design doc) into couple of > smaller ones, hopefully independent. The questions that you have asked > Fabian have for example very little to do with reading metadata from Hive > Meta Store? > > > > > > Piotrek > > > > > >> On 7 Nov 2018, at 14:27, Fabian Hueske <[hidden email]> wrote: > > >> > > >> Hi Xuefu and all, > > >> > > >> Thanks for sharing this design document! > > > > >> I'm very much in favor of restructuring / reworking the catalog > handling in > > >> Flink SQL as outlined in the document. > > > > >> Most changes described in the design document seem to be rather > general and > > >> not specifically related to the Hive integration. > > >> > > > > >> IMO, there are some aspects, especially those at the boundary of Hive > and > > >> Flink, that need a bit more discussion. For example > > >> > > >> * What does it take to make Flink schema compatible with Hive schema? > > >> * How will Flink tables (descriptors) be stored in HMS? > > >> * How do both Hive catalogs differ? Could they be integrated into to a > > >> single one? When to use which one? > > > > >> * What meta information is provided by HMS? What of this can be > leveraged > > >> by Flink? > > >> > > >> Thank you, > > >> Fabian > > >> > > >> Am Fr., 2. Nov. 2018 um 00:31 Uhr schrieb Bowen Li < > [hidden email] > > >: > > >> > > >>> After taking a look at how other discussion threads work, I think > it's > > >>> actually fine just keep our discussion here. It's up to you, Xuefu. > > >>> > > >>> The google doc LGTM. I left some minor comments. > > >>> > > >>> On Thu, Nov 1, 2018 at 10:17 AM Bowen Li <[hidden email]> > wrote: > > >>> > > >>>> Hi all, > > >>>> > > >>>> As Xuefu has published the design doc on google, I agree with > Shuyi's > > > > >>>> suggestion that we probably should start a new email thread like > "[DISCUSS] > > > > >>>> ... Hive integration design ..." on only dev mailing list for > community > > >>>> devs to review. The current thread sends to both dev and user list. > > >>>> > > > > >>>> This email thread is more like validating the general idea and > direction > > > > >>>> with the community, and it's been pretty long and crowded so far. > Since > > > > >>>> everyone is pro for the idea, we can move forward with another > thread to > > >>>> discuss and finalize the design. > > >>>> > > >>>> Thanks, > > >>>> Bowen > > >>>> > > >>>> On Wed, Oct 31, 2018 at 12:16 PM Zhang, Xuefu < > > [hidden email]> > > >>>> wrote: > > >>>> > > >>>>> Hi Shuiyi, > > >>>>> > > > > >>>>> Good idea. Actually the PDF was converted from a google doc. Here > is its > > >>>>> link: > > >>>>> > > >>>>> > > > https://docs.google.com/document/d/1SkppRD_rE3uOKSN-LuZCqn4f7dz0zW5aa6T_hBZq5_o/edit?usp=sharing > > >>>>> Once we reach an agreement, I can convert it to a FLIP. > > >>>>> > > >>>>> Thanks, > > >>>>> Xuefu > > >>>>> > > >>>>> > > >>>>> > > >>>>> ------------------------------------------------------------------ > > >>>>> Sender:Shuyi Chen <[hidden email]> > > >>>>> Sent at:2018 Nov 1 (Thu) 02:47 > > >>>>> Recipient:Xuefu <[hidden email]> > > >>>>> Cc:vino yang <[hidden email]>; Fabian Hueske < > > [hidden email]>; > > >>>>> dev <[hidden email]>; user <[hidden email]> > > >>>>> Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem > > >>>>> > > >>>>> Hi Xuefu, > > >>>>> > > > > >>>>> Thanks a lot for driving this big effort. I would suggest convert > your > > > > >>>>> proposal and design doc into a google doc, and share it on the dev > mailing > > > > >>>>> list for the community to review and comment with title like > "[DISCUSS] ... > > > > >>>>> Hive integration design ..." . Once approved, we can document it > as a FLIP > > > > >>>>> (Flink Improvement Proposals), and use JIRAs to track the > implementations. > > >>>>> What do you think? > > >>>>> > > >>>>> Shuyi > > >>>>> > > >>>>> On Tue, Oct 30, 2018 at 11:32 AM Zhang, Xuefu < > > [hidden email]> > > >>>>> wrote: > > >>>>> Hi all, > > >>>>> > > >>>>> I have also shared a design doc on Hive metastore integration that > is > > > > >>>>> attached here and also to FLINK-10556[1]. Please kindly review and > share > > >>>>> your feedback. > > >>>>> > > >>>>> > > >>>>> Thanks, > > >>>>> Xuefu > > >>>>> > > >>>>> [1] https://issues.apache.org/jira/browse/FLINK-10556 > > >>>>> ------------------------------------------------------------------ > > >>>>> Sender:Xuefu <[hidden email]> > > >>>>> Sent at:2018 Oct 25 (Thu) 01:08 > > >>>>> Recipient:Xuefu <[hidden email]>; Shuyi Chen < > > >>>>> [hidden email]> > > >>>>> Cc:yanghua1127 <[hidden email]>; Fabian Hueske < > > [hidden email]>; > > >>>>> dev <[hidden email]>; user <[hidden email]> > > >>>>> Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem > > >>>>> > > >>>>> Hi all, > > >>>>> > > >>>>> To wrap up the discussion, I have attached a PDF describing the > > > > >>>>> proposal, which is also attached to FLINK-10556 [1]. Please feel > free to > > >>>>> watch that JIRA to track the progress. > > >>>>> > > >>>>> Please also let me know if you have additional comments or > questions. > > >>>>> > > >>>>> Thanks, > > >>>>> Xuefu > > >>>>> > > >>>>> [1] https://issues.apache.org/jira/browse/FLINK-10556 > > >>>>> > > >>>>> > > >>>>> ------------------------------------------------------------------ > > >>>>> Sender:Xuefu <[hidden email]> > > >>>>> Sent at:2018 Oct 16 (Tue) 03:40 > > >>>>> Recipient:Shuyi Chen <[hidden email]> > > >>>>> Cc:yanghua1127 <[hidden email]>; Fabian Hueske < > > [hidden email]>; > > >>>>> dev <[hidden email]>; user <[hidden email]> > > >>>>> Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem > > >>>>> > > >>>>> Hi Shuyi, > > >>>>> > > > > >>>>> Thank you for your input. Yes, I agreed with a phased approach and > like > > > > >>>>> to move forward fast. :) We did some work internally on DDL > utilizing babel > > >>>>> parser in Calcite. While babel makes Calcite's grammar extensible, > at > > >>>>> first impression it still seems too cumbersome for a project when > too > > > > >>>>> much extensions are made. It's even challenging to find where the > extension > > > > >>>>> is needed! It would be certainly better if Calcite can magically > support > > > > >>>>> Hive QL by just turning on a flag, such as that for MYSQL_5. I can > also > > > > >>>>> see that this could mean a lot of work on Calcite. Nevertheless, I > will > > > > >>>>> bring up the discussion over there and to see what their community > thinks. > > >>>>> > > >>>>> Would mind to share more info about the proposal on DDL that you > > >>>>> mentioned? We can certainly collaborate on this. > > >>>>> > > >>>>> Thanks, > > >>>>> Xuefu > > >>>>> > > >>>>> ------------------------------------------------------------------ > > >>>>> Sender:Shuyi Chen <[hidden email]> > > >>>>> Sent at:2018 Oct 14 (Sun) 08:30 > > >>>>> Recipient:Xuefu <[hidden email]> > > >>>>> Cc:yanghua1127 <[hidden email]>; Fabian Hueske < > > [hidden email]>; > > >>>>> dev <[hidden email]>; user <[hidden email]> > > >>>>> Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem > > >>>>> > > >>>>> Welcome to the community and thanks for the great proposal, Xuefu! > I > > > > >>>>> think the proposal can be divided into 2 stages: making Flink to > support > > > > >>>>> Hive features, and make Hive to work with Flink. I agreed with > Timo that on > > > > >>>>> starting with a smaller scope, so we can make progress faster. As > for [6], > > > > >>>>> a proposal for DDL is already in progress, and will come after the > unified > > > > >>>>> SQL connector API is done. For supporting Hive syntax, we might > need to > > >>>>> work with the Calcite community, and a recent effort called babel ( > > >>>>> https://issues.apache.org/jira/browse/CALCITE-2280) in Calcite > might > > >>>>> help here. > > >>>>> > > >>>>> Thanks > > >>>>> Shuyi > > >>>>> > > >>>>> On Wed, Oct 10, 2018 at 8:02 PM Zhang, Xuefu < > > [hidden email]> > > >>>>> wrote: > > >>>>> Hi Fabian/Vno, > > >>>>> > > > > >>>>> Thank you very much for your encouragement inquiry. Sorry that I > didn't > > > > >>>>> see Fabian's email until I read Vino's response just now. (Somehow > Fabian's > > >>>>> went to the spam folder.) > > >>>>> > > > > >>>>> My proposal contains long-term and short-terms goals. > Nevertheless, the > > >>>>> effort will focus on the following areas, including Fabian's list: > > >>>>> > > >>>>> 1. Hive metastore connectivity - This covers both read/write > access, > > > > >>>>> which means Flink can make full use of Hive's metastore as its > catalog (at > > >>>>> least for the batch but can extend for streaming as well). > > > > >>>>> 2. Metadata compatibility - Objects (databases, tables, > partitions, etc) > > > > >>>>> created by Hive can be understood by Flink and the reverse > direction is > > >>>>> true also. > > >>>>> 3. Data compatibility - Similar to #2, data produced by Hive can be > > >>>>> consumed by Flink and vise versa. > > > > >>>>> 4. Support Hive UDFs - For all Hive's native udfs, Flink either > provides > > >>>>> its own implementation or make Hive's implementation work in Flink. > > >>>>> Further, for user created UDFs in Hive, Flink SQL should provide a > > > > >>>>> mechanism allowing user to import them into Flink without any code > change > > >>>>> required. > > >>>>> 5. Data types - Flink SQL should support all data types that are > > >>>>> available in Hive. > > >>>>> 6. SQL Language - Flink SQL should support SQL standard (such as > > > > >>>>> SQL2003) with extension to support Hive's syntax and language > features, > > >>>>> around DDL, DML, and SELECT queries. > > > > >>>>> 7. SQL CLI - this is currently developing in Flink but more > effort is > > >>>>> needed. > > > > >>>>> 8. Server - provide a server that's compatible with Hive's > HiverServer2 > > > > >>>>> in thrift APIs, such that HiveServer2 users can reuse their > existing client > > >>>>> (such as beeline) but connect to Flink's thrift server instead. > > > > >>>>> 9. JDBC/ODBC drivers - Flink may provide its own JDBC/ODBC drivers > for > > >>>>> other application to use to connect to its thrift server > > >>>>> 10. Support other user's customizations in Hive, such as Hive > Serdes, > > >>>>> storage handlers, etc. > > > > >>>>> 11. Better task failure tolerance and task scheduling at Flink > runtime. > > >>>>> > > >>>>> As you can see, achieving all those requires significant effort and > > > > >>>>> across all layers in Flink. However, a short-term goal could > include only > > > > >>>>> core areas (such as 1, 2, 4, 5, 6, 7) or start at a smaller scope > (such as > > >>>>> #3, #6). > > >>>>> > > > > >>>>> Please share your further thoughts. If we generally agree that > this is > > > > >>>>> the right direction, I could come up with a formal proposal > quickly and > > >>>>> then we can follow up with broader discussions. > > >>>>> > > >>>>> Thanks, > > >>>>> Xuefu > > >>>>> > > >>>>> > > >>>>> > > >>>>> ------------------------------------------------------------------ > > >>>>> Sender:vino yang <[hidden email]> > > >>>>> Sent at:2018 Oct 11 (Thu) 09:45 > > >>>>> Recipient:Fabian Hueske <[hidden email]> > > >>>>> Cc:dev <[hidden email]>; Xuefu <[hidden email] > > >; user < > > >>>>> [hidden email]> > > >>>>> Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem > > >>>>> > > >>>>> Hi Xuefu, > > >>>>> > > > > >>>>> Appreciate this proposal, and like Fabian, it would look better if > you > > >>>>> can give more details of the plan. > > >>>>> > > >>>>> Thanks, vino. > > >>>>> > > >>>>> Fabian Hueske <[hidden email]> 于2018年10月10日周三 下午5:27写道： > > >>>>> Hi Xuefu, > > >>>>> > > > > >>>>> Welcome to the Flink community and thanks for starting this > discussion! > > >>>>> Better Hive integration would be really great! > > >>>>> Can you go into details of what you are proposing? I can think of a > > >>>>> couple ways to improve Flink in that regard: > > >>>>> > > >>>>> * Support for Hive UDFs > > >>>>> * Support for Hive metadata catalog > > >>>>> * Support for HiveQL syntax > > >>>>> * ??? > > >>>>> > > >>>>> Best, Fabian > > >>>>> > > >>>>> Am Di., 9. Okt. 2018 um 19:22 Uhr schrieb Zhang, Xuefu < > > >>>>> [hidden email]>: > > >>>>> Hi all, > > >>>>> > > >>>>> Along with the community's effort, inside Alibaba we have explored > > > > >>>>> Flink's potential as an execution engine not just for stream > processing but > > >>>>> also for batch processing. We are encouraged by our findings and > have > > > > >>>>> initiated our effort to make Flink's SQL capabilities > full-fledged. When > > > > >>>>> comparing what's available in Flink to the offerings from > competitive data > > > > >>>>> processing engines, we identified a major gap in Flink: a well > integration > > > > >>>>> with Hive ecosystem. This is crucial to the success of Flink SQL > and batch > > > > >>>>> due to the well-established data ecosystem around Hive. Therefore, > we have > > > > >>>>> done some initial work along this direction but there are still a > lot of > > >>>>> effort needed. > > >>>>> > > >>>>> We have two strategies in mind. The first one is to make Flink SQL > > > > >>>>> full-fledged and well-integrated with Hive ecosystem. This is a > similar > > > > >>>>> approach to what Spark SQL adopted. The second strategy is to make > Hive > > > > >>>>> itself work with Flink, similar to the proposal in [1]. Each > approach bears > > > > >>>>> its pros and cons, but they don’t need to be mutually exclusive > with each > > >>>>> targeting at different users and use cases. We believe that both > will > > >>>>> promote a much greater adoption of Flink beyond stream processing. > > >>>>> > > >>>>> We have been focused on the first approach and would like to > showcase > > > > >>>>> Flink's batch and SQL capabilities with Flink SQL. However, we > have also > > >>>>> planned to start strategy #2 as the follow-up effort. > > >>>>> > > > > >>>>> I'm completely new to Flink(, with a short bio [2] below), though > many > > > > >>>>> of my colleagues here at Alibaba are long-time contributors. > Nevertheless, > > > > >>>>> I'd like to share our thoughts and invite your early feedback. At > the same > > > > >>>>> time, I am working on a detailed proposal on Flink SQL's > integration with > > >>>>> Hive ecosystem, which will be also shared when ready. > > >>>>> > > >>>>> While the ideas are simple, each approach will demand significant > > > > >>>>> effort, more than what we can afford. Thus, the input and > contributions > > >>>>> from the communities are greatly welcome and appreciated. > > >>>>> > > >>>>> Regards, > > >>>>> > > >>>>> > > >>>>> Xuefu > > >>>>> > > >>>>> References: > > >>>>> > > >>>>> [1] https://issues.apache.org/jira/browse/HIVE-10712 > > > > >>>>> [2] Xuefu Zhang is a long-time open source veteran, worked or > working on > > >>>>> many projects under Apache Foundation, of which he is also an > honored > > > > >>>>> member. About 10 years ago he worked in the Hadoop team at Yahoo > where the > > > > >>>>> projects just got started. Later he worked at Cloudera, initiating > and > > > > >>>>> leading the development of Hive on Spark project in the > communities and > > > > >>>>> across many organizations. Prior to joining Alibaba, he worked at > Uber > > > > >>>>> where he promoted Hive on Spark to all Uber's SQL on Hadoop > workload and > > >>>>> significantly improved Uber's cluster efficiency. > > >>>>> > > >>>>> > > >>>>> > > >>>>> > > >>>>> -- > > > > >>>>> "So you have to trust that the dots will somehow connect in your > future." > > >>>>> > > >>>>> > > >>>>> -- > > > > >>>>> "So you have to trust that the dots will somehow connect in your > future." > > >>>>> > > > > > |
Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem
|
|
Hi Xiaowei,
Thanks for bringing up the question. In the current design, the properties for meta objects are meant to cover anything that's specific to a particular catalog and agnostic to Flink. Anything that is common (such as schema for tables, query text for views, and udf classname) are abstracted as members of the respective classes. However, this is still in discussion, and Timo and I will go over this and provide an update. Please note that UDF is a little more involved than what the current design doc shows. I'm still refining this part. Thanks, Xuefu ------------------------------------------------------------------ Sender:Xiaowei Jiang <[hidden email]> Sent at:2018 Nov 18 (Sun) 15:17 Recipient:dev <[hidden email]> Cc:Xuefu <[hidden email]>; twalthr <[hidden email]>; piotr <[hidden email]>; Fabian Hueske <[hidden email]>; suez1224 <[hidden email]> Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem Thanks Xuefu for the detailed design doc! One question on the properties associated with the catalog objects. Are we going to leave them completely free form or we are going to set some standard for that? I think that the answer may depend on if we want to explore catalog specific optimization opportunities. In any case, I think that it might be helpful for standardize as much as possible into strongly typed classes and use leave these properties for catalog specific things. But I think that we can do it in steps. Xiaowei On Fri, Nov 16, 2018 at 4:00 AM Bowen Li <[hidden email]> wrote: Thanks for keeping on improving the overall design, Xuefu! It looks quite good to me now. Would be nice that cc-ed Flink committers can help to review and confirm! One minor suggestion: Since the last section of design doc already touches some new sql statements, shall we add another section in our doc and formalize the new sql statements in SQL Client and TableEnvironment that are gonna come along naturally with our design? Here are some that the design doc mentioned and some that I came up with: To be added: - USE <catalog> - set default catalog - USE <catalog.schema> - set default schema - SHOW CATALOGS - show all registered catalogs - SHOW SCHEMAS [FROM catalog] - list schemas in the current default catalog or the specified catalog - DESCRIBE VIEW view - show the view's definition in CatalogView - SHOW VIEWS [FROM schema/catalog.schema] - show views from current or a specified schema. (DDLs that can be addressed by either our design or Shuyi's DDL design) - CREATE/DROP/ALTER SCHEMA schema - CREATE/DROP/ALTER CATALOG catalog To be modified: - SHOW TABLES [FROM schema/catalog.schema] - show tables from current or a specified schema. Add 'from schema' to existing 'SHOW TABLES' statement - SHOW FUNCTIONS [FROM schema/catalog.schema] - show functions from current or a specified schema. Add 'from schema' to existing 'SHOW TABLES' statement' Thanks, Bowen On Wed, Nov 14, 2018 at 10:39 PM Zhang, Xuefu <[hidden email]> wrote: > Thanks, Bowen, for catching the error. I have granted comment permission > with the link. > > I also updated the doc with the latest class definitions. Everyone is > encouraged to review and comment. > > Thanks, > Xuefu > > ------------------------------------------------------------------ > Sender:Bowen Li <[hidden email]> > Sent at:2018 Nov 14 (Wed) 06:44 > Recipient:Xuefu <[hidden email]> > Cc:piotr <[hidden email]>; dev <[hidden email]>; Shuyi > Chen <[hidden email]> > Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem > > Hi Xuefu, > > Currently the new design doc > <https://docs.google.com/document/d/1Y9it78yaUvbv4g572ZK_lZnZaAGjqwM_EhjdOv4yJtw/edit> > is on “view only" mode, and people cannot leave comments. Can you please > change it to "can comment" or "can edit" mode? > > Thanks, Bowen > > > On Mon, Nov 12, 2018 at 9:51 PM Zhang, Xuefu <[hidden email]> > wrote: > Hi Piotr > > I have extracted the API portion of the design and the google doc is here > <https://docs.google.com/document/d/1Y9it78yaUvbv4g572ZK_lZnZaAGjqwM_EhjdOv4yJtw/edit?usp=sharing>. > Please review and provide your feedback. > > Thanks, > Xuefu > > ------------------------------------------------------------------ > Sender:Xuefu <[hidden email]> > Sent at:2018 Nov 12 (Mon) 12:43 > Recipient:Piotr Nowojski <[hidden email]>; dev < > [hidden email]> > Cc:Bowen Li <[hidden email]>; Shuyi Chen <[hidden email]> > Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem > > Hi Piotr, > > That sounds good to me. Let's close all the open questions ((there are a > couple of them)) in the Google doc and I should be able to quickly split > it into the three proposals as you suggested. > > Thanks, > Xuefu > > ------------------------------------------------------------------ > Sender:Piotr Nowojski <[hidden email]> > Sent at:2018 Nov 9 (Fri) 22:46 > Recipient:dev <[hidden email]>; Xuefu <[hidden email]> > Cc:Bowen Li <[hidden email]>; Shuyi Chen <[hidden email]> > Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem > > Hi, > > > Yes, it seems like the best solution. Maybe someone else can also suggests if we can split it further? Maybe changes in the interface in one doc, reading from hive meta store another and final storing our meta informations in hive meta store? > > Piotrek > > > On 9 Nov 2018, at 01:44, Zhang, Xuefu <[hidden email]> wrote: > > > > Hi Piotr, > > > > That seems to be good idea! > > > > > Since the google doc for the design is currently under extensive review, I will leave it as it is for now. However, I'll convert it to two different FLIPs when the time comes. > > > > How does it sound to you? > > > > Thanks, > > Xuefu > > > > > > ------------------------------------------------------------------ > > Sender:Piotr Nowojski <[hidden email]> > > Sent at:2018 Nov 9 (Fri) 02:31 > > Recipient:dev <[hidden email]> > > Cc:Bowen Li <[hidden email]>; Xuefu <[hidden email] > >; Shuyi Chen <[hidden email]> > > Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem > > > > Hi, > > > > > Maybe we should split this topic (and the design doc) into couple of smaller ones, hopefully independent. The questions that you have asked Fabian have for example very little to do with reading metadata from Hive Meta Store? > > > > Piotrek > > > >> On 7 Nov 2018, at 14:27, Fabian Hueske <[hidden email]> wrote: > >> > >> Hi Xuefu and all, > >> > >> Thanks for sharing this design document! > > >> I'm very much in favor of restructuring / reworking the catalog handling in > >> Flink SQL as outlined in the document. > > >> Most changes described in the design document seem to be rather general and > >> not specifically related to the Hive integration. > >> > > >> IMO, there are some aspects, especially those at the boundary of Hive and > >> Flink, that need a bit more discussion. For example > >> > >> * What does it take to make Flink schema compatible with Hive schema? > >> * How will Flink tables (descriptors) be stored in HMS? > >> * How do both Hive catalogs differ? Could they be integrated into to a > >> single one? When to use which one? > > >> * What meta information is provided by HMS? What of this can be leveraged > >> by Flink? > >> > >> Thank you, > >> Fabian > >> > >> Am Fr., 2. Nov. 2018 um 00:31 Uhr schrieb Bowen Li <[hidden email] > >: > >> > >>> After taking a look at how other discussion threads work, I think it's > >>> actually fine just keep our discussion here. It's up to you, Xuefu. > >>> > >>> The google doc LGTM. I left some minor comments. > >>> > >>> On Thu, Nov 1, 2018 at 10:17 AM Bowen Li <[hidden email]> wrote: > >>> > >>>> Hi all, > >>>> > >>>> As Xuefu has published the design doc on google, I agree with Shuyi's > > >>>> suggestion that we probably should start a new email thread like "[DISCUSS] > > >>>> ... Hive integration design ..." on only dev mailing list for community > >>>> devs to review. The current thread sends to both dev and user list. > >>>> > > >>>> This email thread is more like validating the general idea and direction > > >>>> with the community, and it's been pretty long and crowded so far. Since > > >>>> everyone is pro for the idea, we can move forward with another thread to > >>>> discuss and finalize the design. > >>>> > >>>> Thanks, > >>>> Bowen > >>>> > >>>> On Wed, Oct 31, 2018 at 12:16 PM Zhang, Xuefu < > [hidden email]> > >>>> wrote: > >>>> > >>>>> Hi Shuiyi, > >>>>> > > >>>>> Good idea. Actually the PDF was converted from a google doc. Here is its > >>>>> link: > >>>>> > >>>>> > https://docs.google.com/document/d/1SkppRD_rE3uOKSN-LuZCqn4f7dz0zW5aa6T_hBZq5_o/edit?usp=sharing > >>>>> Once we reach an agreement, I can convert it to a FLIP. > >>>>> > >>>>> Thanks, > >>>>> Xuefu > >>>>> > >>>>> > >>>>> > >>>>> ------------------------------------------------------------------ > >>>>> Sender:Shuyi Chen <[hidden email]> > >>>>> Sent at:2018 Nov 1 (Thu) 02:47 > >>>>> Recipient:Xuefu <[hidden email]> > >>>>> Cc:vino yang <[hidden email]>; Fabian Hueske < > [hidden email]>; > >>>>> dev <[hidden email]>; user <[hidden email]> > >>>>> Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem > >>>>> > >>>>> Hi Xuefu, > >>>>> > > >>>>> Thanks a lot for driving this big effort. I would suggest convert your > > >>>>> proposal and design doc into a google doc, and share it on the dev mailing > > >>>>> list for the community to review and comment with title like "[DISCUSS] ... > > >>>>> Hive integration design ..." . Once approved, we can document it as a FLIP > > >>>>> (Flink Improvement Proposals), and use JIRAs to track the implementations. > >>>>> What do you think? > >>>>> > >>>>> Shuyi > >>>>> > >>>>> On Tue, Oct 30, 2018 at 11:32 AM Zhang, Xuefu < > [hidden email]> > >>>>> wrote: > >>>>> Hi all, > >>>>> > >>>>> I have also shared a design doc on Hive metastore integration that is > > >>>>> attached here and also to FLINK-10556[1]. Please kindly review and share > >>>>> your feedback. > >>>>> > >>>>> > >>>>> Thanks, > >>>>> Xuefu > >>>>> > >>>>> [1] https://issues.apache.org/jira/browse/FLINK-10556 > >>>>> ------------------------------------------------------------------ > >>>>> Sender:Xuefu <[hidden email]> > >>>>> Sent at:2018 Oct 25 (Thu) 01:08 > >>>>> Recipient:Xuefu <[hidden email]>; Shuyi Chen < > >>>>> [hidden email]> > >>>>> Cc:yanghua1127 <[hidden email]>; Fabian Hueske < > [hidden email]>; > >>>>> dev <[hidden email]>; user <[hidden email]> > >>>>> Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem > >>>>> > >>>>> Hi all, > >>>>> > >>>>> To wrap up the discussion, I have attached a PDF describing the > > >>>>> proposal, which is also attached to FLINK-10556 [1]. Please feel free to > >>>>> watch that JIRA to track the progress. > >>>>> > >>>>> Please also let me know if you have additional comments or questions. > >>>>> > >>>>> Thanks, > >>>>> Xuefu > >>>>> > >>>>> [1] https://issues.apache.org/jira/browse/FLINK-10556 > >>>>> > >>>>> > >>>>> ------------------------------------------------------------------ > >>>>> Sender:Xuefu <[hidden email]> > >>>>> Sent at:2018 Oct 16 (Tue) 03:40 > >>>>> Recipient:Shuyi Chen <[hidden email]> > >>>>> Cc:yanghua1127 <[hidden email]>; Fabian Hueske < > [hidden email]>; > >>>>> dev <[hidden email]>; user <[hidden email]> > >>>>> Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem > >>>>> > >>>>> Hi Shuyi, > >>>>> > > >>>>> Thank you for your input. Yes, I agreed with a phased approach and like > > >>>>> to move forward fast. :) We did some work internally on DDL utilizing babel > >>>>> parser in Calcite. While babel makes Calcite's grammar extensible, at > >>>>> first impression it still seems too cumbersome for a project when too > > >>>>> much extensions are made. It's even challenging to find where the extension > > >>>>> is needed! It would be certainly better if Calcite can magically support > > >>>>> Hive QL by just turning on a flag, such as that for MYSQL_5. I can also > > >>>>> see that this could mean a lot of work on Calcite. Nevertheless, I will > > >>>>> bring up the discussion over there and to see what their community thinks. > >>>>> > >>>>> Would mind to share more info about the proposal on DDL that you > >>>>> mentioned? We can certainly collaborate on this. > >>>>> > >>>>> Thanks, > >>>>> Xuefu > >>>>> > >>>>> ------------------------------------------------------------------ > >>>>> Sender:Shuyi Chen <[hidden email]> > >>>>> Sent at:2018 Oct 14 (Sun) 08:30 > >>>>> Recipient:Xuefu <[hidden email]> > >>>>> Cc:yanghua1127 <[hidden email]>; Fabian Hueske < > [hidden email]>; > >>>>> dev <[hidden email]>; user <[hidden email]> > >>>>> Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem > >>>>> > >>>>> Welcome to the community and thanks for the great proposal, Xuefu! I > > >>>>> think the proposal can be divided into 2 stages: making Flink to support > > >>>>> Hive features, and make Hive to work with Flink. I agreed with Timo that on > > >>>>> starting with a smaller scope, so we can make progress faster. As for [6], > > >>>>> a proposal for DDL is already in progress, and will come after the unified > > >>>>> SQL connector API is done. For supporting Hive syntax, we might need to > >>>>> work with the Calcite community, and a recent effort called babel ( > >>>>> https://issues.apache.org/jira/browse/CALCITE-2280) in Calcite might > >>>>> help here. > >>>>> > >>>>> Thanks > >>>>> Shuyi > >>>>> > >>>>> On Wed, Oct 10, 2018 at 8:02 PM Zhang, Xuefu < > [hidden email]> > >>>>> wrote: > >>>>> Hi Fabian/Vno, > >>>>> > > >>>>> Thank you very much for your encouragement inquiry. Sorry that I didn't > > >>>>> see Fabian's email until I read Vino's response just now. (Somehow Fabian's > >>>>> went to the spam folder.) > >>>>> > > >>>>> My proposal contains long-term and short-terms goals. Nevertheless, the > >>>>> effort will focus on the following areas, including Fabian's list: > >>>>> > >>>>> 1. Hive metastore connectivity - This covers both read/write access, > > >>>>> which means Flink can make full use of Hive's metastore as its catalog (at > >>>>> least for the batch but can extend for streaming as well). > > >>>>> 2. Metadata compatibility - Objects (databases, tables, partitions, etc) > > >>>>> created by Hive can be understood by Flink and the reverse direction is > >>>>> true also. > >>>>> 3. Data compatibility - Similar to #2, data produced by Hive can be > >>>>> consumed by Flink and vise versa. > > >>>>> 4. Support Hive UDFs - For all Hive's native udfs, Flink either provides > >>>>> its own implementation or make Hive's implementation work in Flink. > >>>>> Further, for user created UDFs in Hive, Flink SQL should provide a > > >>>>> mechanism allowing user to import them into Flink without any code change > >>>>> required. > >>>>> 5. Data types - Flink SQL should support all data types that are > >>>>> available in Hive. > >>>>> 6. SQL Language - Flink SQL should support SQL standard (such as > > >>>>> SQL2003) with extension to support Hive's syntax and language features, > >>>>> around DDL, DML, and SELECT queries. > > >>>>> 7. SQL CLI - this is currently developing in Flink but more effort is > >>>>> needed. > > >>>>> 8. Server - provide a server that's compatible with Hive's HiverServer2 > > >>>>> in thrift APIs, such that HiveServer2 users can reuse their existing client > >>>>> (such as beeline) but connect to Flink's thrift server instead. > > >>>>> 9. JDBC/ODBC drivers - Flink may provide its own JDBC/ODBC drivers for > >>>>> other application to use to connect to its thrift server > >>>>> 10. Support other user's customizations in Hive, such as Hive Serdes, > >>>>> storage handlers, etc. > > >>>>> 11. Better task failure tolerance and task scheduling at Flink runtime. > >>>>> > >>>>> As you can see, achieving all those requires significant effort and > > >>>>> across all layers in Flink. However, a short-term goal could include only > > >>>>> core areas (such as 1, 2, 4, 5, 6, 7) or start at a smaller scope (such as > >>>>> #3, #6). > >>>>> > > >>>>> Please share your further thoughts. If we generally agree that this is > > >>>>> the right direction, I could come up with a formal proposal quickly and > >>>>> then we can follow up with broader discussions. > >>>>> > >>>>> Thanks, > >>>>> Xuefu > >>>>> > >>>>> > >>>>> > >>>>> ------------------------------------------------------------------ > >>>>> Sender:vino yang <[hidden email]> > >>>>> Sent at:2018 Oct 11 (Thu) 09:45 > >>>>> Recipient:Fabian Hueske <[hidden email]> > >>>>> Cc:dev <[hidden email]>; Xuefu <[hidden email] > >; user < > >>>>> [hidden email]> > >>>>> Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem > >>>>> > >>>>> Hi Xuefu, > >>>>> > > >>>>> Appreciate this proposal, and like Fabian, it would look better if you > >>>>> can give more details of the plan. > >>>>> > >>>>> Thanks, vino. > >>>>> > >>>>> Fabian Hueske <[hidden email]> 于2018年10月10日周三 下午5:27写道： > >>>>> Hi Xuefu, > >>>>> > > >>>>> Welcome to the Flink community and thanks for starting this discussion! > >>>>> Better Hive integration would be really great! > >>>>> Can you go into details of what you are proposing? I can think of a > >>>>> couple ways to improve Flink in that regard: > >>>>> > >>>>> * Support for Hive UDFs > >>>>> * Support for Hive metadata catalog > >>>>> * Support for HiveQL syntax > >>>>> * ??? > >>>>> > >>>>> Best, Fabian > >>>>> > >>>>> Am Di., 9. Okt. 2018 um 19:22 Uhr schrieb Zhang, Xuefu < > >>>>> [hidden email]>: > >>>>> Hi all, > >>>>> > >>>>> Along with the community's effort, inside Alibaba we have explored > > >>>>> Flink's potential as an execution engine not just for stream processing but > >>>>> also for batch processing. We are encouraged by our findings and have > > >>>>> initiated our effort to make Flink's SQL capabilities full-fledged. When > > >>>>> comparing what's available in Flink to the offerings from competitive data > > >>>>> processing engines, we identified a major gap in Flink: a well integration > > >>>>> with Hive ecosystem. This is crucial to the success of Flink SQL and batch > > >>>>> due to the well-established data ecosystem around Hive. Therefore, we have > > >>>>> done some initial work along this direction but there are still a lot of > >>>>> effort needed. > >>>>> > >>>>> We have two strategies in mind. The first one is to make Flink SQL > > >>>>> full-fledged and well-integrated with Hive ecosystem. This is a similar > > >>>>> approach to what Spark SQL adopted. The second strategy is to make Hive > > >>>>> itself work with Flink, similar to the proposal in [1]. Each approach bears > > >>>>> its pros and cons, but they don’t need to be mutually exclusive with each > >>>>> targeting at different users and use cases. We believe that both will > >>>>> promote a much greater adoption of Flink beyond stream processing. > >>>>> > >>>>> We have been focused on the first approach and would like to showcase > > >>>>> Flink's batch and SQL capabilities with Flink SQL. However, we have also > >>>>> planned to start strategy #2 as the follow-up effort. > >>>>> > > >>>>> I'm completely new to Flink(, with a short bio [2] below), though many > > >>>>> of my colleagues here at Alibaba are long-time contributors. Nevertheless, > > >>>>> I'd like to share our thoughts and invite your early feedback. At the same > > >>>>> time, I am working on a detailed proposal on Flink SQL's integration with > >>>>> Hive ecosystem, which will be also shared when ready. > >>>>> > >>>>> While the ideas are simple, each approach will demand significant > > >>>>> effort, more than what we can afford. Thus, the input and contributions > >>>>> from the communities are greatly welcome and appreciated. > >>>>> > >>>>> Regards, > >>>>> > >>>>> > >>>>> Xuefu > >>>>> > >>>>> References: > >>>>> > >>>>> [1] https://issues.apache.org/jira/browse/HIVE-10712 > > >>>>> [2] Xuefu Zhang is a long-time open source veteran, worked or working on > >>>>> many projects under Apache Foundation, of which he is also an honored > > >>>>> member. About 10 years ago he worked in the Hadoop team at Yahoo where the > > >>>>> projects just got started. Later he worked at Cloudera, initiating and > > >>>>> leading the development of Hive on Spark project in the communities and > > >>>>> across many organizations. Prior to joining Alibaba, he worked at Uber > > >>>>> where he promoted Hive on Spark to all Uber's SQL on Hadoop workload and > >>>>> significantly improved Uber's cluster efficiency. > >>>>> > >>>>> > >>>>> > >>>>> > >>>>> -- > > >>>>> "So you have to trust that the dots will somehow connect in your future." > >>>>> > >>>>> > >>>>> -- > > >>>>> "So you have to trust that the dots will somehow connect in your future." > >>>>> > > |
Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem
|
|
Hi folks, there's clearly some incremental steps to be taken to introduce
catalog support to SQL Client, complementary to what is proposed in the Flink-Hive Metastore design doc. I was quietly working on this over the holidays. I posted some new sub-tasks, PRs, and sample code to FLINK-10744. What inspired me to get involved is that the catalog interface seems like a great way to encapsulate a 'library' of Flink tables and functions. For example, the NYC Taxi dataset (TaxiRides, TaxiFares, various UDFs) may be nicely encapsulated as a catalog (TaxiData). Such a library should be fully consumable in SQL Client. I implemented the above. Some highlights: 1. A fully-worked example of using the Taxi dataset in SQL Client via an environment file. - an ASCII video showing the SQL Client in action: https://asciinema.org/a/C8xuAjmZSxCuApgFgZQyeIHuo - the corresponding environment file (will be even more concise once 'FLINK-10696 Catalog UDFs' is merged): *https://github.com/EronWright/flink-training-exercises/blob/3be008d64be975ced0f1a7e3901a8c5353f72a7e/src/main/dist/conf/sql-client-defaults.yaml <https://github.com/EronWright/flink-training-exercises/blob/3be008d64be975ced0f1a7e3901a8c5353f72a7e/src/main/dist/conf/sql-client-defaults.yaml>* - the typed API for standalone table applications: *https://github.com/EronWright/flink-training-exercises/blob/3be008d64be975ced0f1a7e3901a8c5353f72a7e/src/main/java/com/dataartisans/flinktraining/examples/table_java/examples/ViaCatalog.java#L50 <https://github.com/EronWright/flink-training-exercises/blob/3be008d64be975ced0f1a7e3901a8c5353f72a7e/src/main/java/com/dataartisans/flinktraining/examples/table_java/examples/ViaCatalog.java#L50>* 2. Implementation of the core catalog descriptor and factory. I realize that some renames may later occur as per the design doc, and would be happy to do that as a follow-up. https://github.com/apache/flink/pull/7390 3. Implementation of a connect-style API on TableEnvironment to use catalog descriptor. https://github.com/apache/flink/pull/7392 4. Integration into SQL-Client's environment file: https://github.com/apache/flink/pull/7393 I realize that the overall Hive integration is still evolving, but I believe that these PRs are a good stepping stone. Here's the list (in bottom-up order): - https://github.com/apache/flink/pull/7386 - https://github.com/apache/flink/pull/7388 - https://github.com/apache/flink/pull/7389 - https://github.com/apache/flink/pull/7390 - https://github.com/apache/flink/pull/7392 - https://github.com/apache/flink/pull/7393 Thanks and enjoy 2019! Eron W On Sun, Nov 18, 2018 at 3:04 PM Zhang, Xuefu <[hidden email]> wrote: > Hi Xiaowei, > > Thanks for bringing up the question. In the current design, the properties > for meta objects are meant to cover anything that's specific to a > particular catalog and agnostic to Flink. Anything that is common (such as > schema for tables, query text for views, and udf classname) are abstracted > as members of the respective classes. However, this is still in discussion, > and Timo and I will go over this and provide an update. > > Please note that UDF is a little more involved than what the current > design doc shows. I'm still refining this part. > > Thanks, > Xuefu > > > ------------------------------------------------------------------ > Sender:Xiaowei Jiang <[hidden email]> > Sent at:2018 Nov 18 (Sun) 15:17 > Recipient:dev <[hidden email]> > Cc:Xuefu <[hidden email]>; twalthr <[hidden email]>; piotr < > [hidden email]>; Fabian Hueske <[hidden email]>; suez1224 < > [hidden email]> > Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem > > Thanks Xuefu for the detailed design doc! One question on the properties > associated with the catalog objects. Are we going to leave them completely > free form or we are going to set some standard for that? I think that the > answer may depend on if we want to explore catalog specific optimization > opportunities. In any case, I think that it might be helpful for > standardize as much as possible into strongly typed classes and use leave > these properties for catalog specific things. But I think that we can do it > in steps. > > Xiaowei > On Fri, Nov 16, 2018 at 4:00 AM Bowen Li <[hidden email]> wrote: > Thanks for keeping on improving the overall design, Xuefu! It looks quite > good to me now. > > Would be nice that cc-ed Flink committers can help to review and confirm! > > > > One minor suggestion: Since the last section of design doc already touches > some new sql statements, shall we add another section in our doc and > formalize the new sql statements in SQL Client and TableEnvironment that > are gonna come along naturally with our design? Here are some that the > design doc mentioned and some that I came up with: > > To be added: > > - USE <catalog> - set default catalog > - USE <catalog.schema> - set default schema > - SHOW CATALOGS - show all registered catalogs > - SHOW SCHEMAS [FROM catalog] - list schemas in the current default > catalog or the specified catalog > - DESCRIBE VIEW view - show the view's definition in CatalogView > - SHOW VIEWS [FROM schema/catalog.schema] - show views from current or > a > specified schema. > > (DDLs that can be addressed by either our design or Shuyi's DDL design) > > - CREATE/DROP/ALTER SCHEMA schema > - CREATE/DROP/ALTER CATALOG catalog > > To be modified: > > - SHOW TABLES [FROM schema/catalog.schema] - show tables from current > or > a specified schema. Add 'from schema' to existing 'SHOW TABLES' > statement > - SHOW FUNCTIONS [FROM schema/catalog.schema] - show functions from > current or a specified schema. Add 'from schema' to existing 'SHOW > TABLES' > statement' > > > Thanks, Bowen > > > > On Wed, Nov 14, 2018 at 10:39 PM Zhang, Xuefu <[hidden email]> > wrote: > > > Thanks, Bowen, for catching the error. I have granted comment permission > > with the link. > > > > I also updated the doc with the latest class definitions. Everyone is > > encouraged to review and comment. > > > > Thanks, > > Xuefu > > > > ------------------------------------------------------------------ > > Sender:Bowen Li <[hidden email]> > > Sent at:2018 Nov 14 (Wed) 06:44 > > Recipient:Xuefu <[hidden email]> > > Cc:piotr <[hidden email]>; dev <[hidden email]>; Shuyi > > Chen <[hidden email]> > > Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem > > > > Hi Xuefu, > > > > Currently the new design doc > > < > https://docs.google.com/document/d/1Y9it78yaUvbv4g572ZK_lZnZaAGjqwM_EhjdOv4yJtw/edit > > > > is on “view only" mode, and people cannot leave comments. Can you please > > change it to "can comment" or "can edit" mode? > > > > Thanks, Bowen > > > > > > On Mon, Nov 12, 2018 at 9:51 PM Zhang, Xuefu <[hidden email]> > > wrote: > > Hi Piotr > > > > I have extracted the API portion of the design and the google doc is > here > > < > https://docs.google.com/document/d/1Y9it78yaUvbv4g572ZK_lZnZaAGjqwM_EhjdOv4yJtw/edit?usp=sharing > >. > > Please review and provide your feedback. > > > > Thanks, > > Xuefu > > > > ------------------------------------------------------------------ > > Sender:Xuefu <[hidden email]> > > Sent at:2018 Nov 12 (Mon) 12:43 > > Recipient:Piotr Nowojski <[hidden email]>; dev < > > [hidden email]> > > Cc:Bowen Li <[hidden email]>; Shuyi Chen <[hidden email]> > > Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem > > > > Hi Piotr, > > > > That sounds good to me. Let's close all the open questions ((there are a > > couple of them)) in the Google doc and I should be able to quickly split > > it into the three proposals as you suggested. > > > > Thanks, > > Xuefu > > > > ------------------------------------------------------------------ > > Sender:Piotr Nowojski <[hidden email]> > > Sent at:2018 Nov 9 (Fri) 22:46 > > Recipient:dev <[hidden email]>; Xuefu <[hidden email]> > > Cc:Bowen Li <[hidden email]>; Shuyi Chen <[hidden email]> > > Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem > > > > Hi, > > > > > > Yes, it seems like the best solution. Maybe someone else can also > suggests if we can split it further? Maybe changes in the interface in one > doc, reading from hive meta store another and final storing our meta > informations in hive meta store? > > > > Piotrek > > > > > On 9 Nov 2018, at 01:44, Zhang, Xuefu <[hidden email]> > wrote: > > > > > > Hi Piotr, > > > > > > That seems to be good idea! > > > > > > > > Since the google doc for the design is currently under extensive > review, I will leave it as it is for now. However, I'll convert it to two > different FLIPs when the time comes. > > > > > > How does it sound to you? > > > > > > Thanks, > > > Xuefu > > > > > > > > > ------------------------------------------------------------------ > > > Sender:Piotr Nowojski <[hidden email]> > > > Sent at:2018 Nov 9 (Fri) 02:31 > > > Recipient:dev <[hidden email]> > > > Cc:Bowen Li <[hidden email]>; Xuefu <[hidden email] > > >; Shuyi Chen <[hidden email]> > > > Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem > > > > > > Hi, > > > > > > > > Maybe we should split this topic (and the design doc) into couple of > smaller ones, hopefully independent. The questions that you have asked > Fabian have for example very little to do with reading metadata from Hive > Meta Store? > > > > > > Piotrek > > > > > >> On 7 Nov 2018, at 14:27, Fabian Hueske <[hidden email]> wrote: > > >> > > >> Hi Xuefu and all, > > >> > > >> Thanks for sharing this design document! > > > > >> I'm very much in favor of restructuring / reworking the catalog > handling in > > >> Flink SQL as outlined in the document. > > > > >> Most changes described in the design document seem to be rather > general and > > >> not specifically related to the Hive integration. > > >> > > > > >> IMO, there are some aspects, especially those at the boundary of > Hive and > > >> Flink, that need a bit more discussion. For example > > >> > > >> * What does it take to make Flink schema compatible with Hive schema? > > >> * How will Flink tables (descriptors) be stored in HMS? > > >> * How do both Hive catalogs differ? Could they be integrated into to > a > > >> single one? When to use which one? > > > > >> * What meta information is provided by HMS? What of this can be > leveraged > > >> by Flink? > > >> > > >> Thank you, > > >> Fabian > > >> > > >> Am Fr., 2. Nov. 2018 um 00:31 Uhr schrieb Bowen Li < > [hidden email] > > >: > > >> > > >>> After taking a look at how other discussion threads work, I think > it's > > >>> actually fine just keep our discussion here. It's up to you, Xuefu. > > >>> > > >>> The google doc LGTM. I left some minor comments. > > >>> > > >>> On Thu, Nov 1, 2018 at 10:17 AM Bowen Li <[hidden email]> > wrote: > > >>> > > >>>> Hi all, > > >>>> > > >>>> As Xuefu has published the design doc on google, I agree with > Shuyi's > > > > >>>> suggestion that we probably should start a new email thread like > "[DISCUSS] > > > > >>>> ... Hive integration design ..." on only dev mailing list for > community > > >>>> devs to review. The current thread sends to both dev and user list. > > >>>> > > > > >>>> This email thread is more like validating the general idea and > direction > > > > >>>> with the community, and it's been pretty long and crowded so far. > Since > > > > >>>> everyone is pro for the idea, we can move forward with another > thread to > > >>>> discuss and finalize the design. > > >>>> > > >>>> Thanks, > > >>>> Bowen > > >>>> > > >>>> On Wed, Oct 31, 2018 at 12:16 PM Zhang, Xuefu < > > [hidden email]> > > >>>> wrote: > > >>>> > > >>>>> Hi Shuiyi, > > >>>>> > > > > >>>>> Good idea. Actually the PDF was converted from a google doc. Here > is its > > >>>>> link: > > >>>>> > > >>>>> > > > https://docs.google.com/document/d/1SkppRD_rE3uOKSN-LuZCqn4f7dz0zW5aa6T_hBZq5_o/edit?usp=sharing > > >>>>> Once we reach an agreement, I can convert it to a FLIP. > > >>>>> > > >>>>> Thanks, > > >>>>> Xuefu > > >>>>> > > >>>>> > > >>>>> > > >>>>> ------------------------------------------------------------------ > > >>>>> Sender:Shuyi Chen <[hidden email]> > > >>>>> Sent at:2018 Nov 1 (Thu) 02:47 > > >>>>> Recipient:Xuefu <[hidden email]> > > >>>>> Cc:vino yang <[hidden email]>; Fabian Hueske < > > [hidden email]>; > > >>>>> dev <[hidden email]>; user <[hidden email]> > > >>>>> Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem > > >>>>> > > >>>>> Hi Xuefu, > > >>>>> > > > > >>>>> Thanks a lot for driving this big effort. I would suggest convert > your > > > > >>>>> proposal and design doc into a google doc, and share it on the > dev mailing > > > > >>>>> list for the community to review and comment with title like > "[DISCUSS] ... > > > > >>>>> Hive integration design ..." . Once approved, we can document it > as a FLIP > > > > >>>>> (Flink Improvement Proposals), and use JIRAs to track the > implementations. > > >>>>> What do you think? > > >>>>> > > >>>>> Shuyi > > >>>>> > > >>>>> On Tue, Oct 30, 2018 at 11:32 AM Zhang, Xuefu < > > [hidden email]> > > >>>>> wrote: > > >>>>> Hi all, > > >>>>> > > >>>>> I have also shared a design doc on Hive metastore integration > that is > > > > >>>>> attached here and also to FLINK-10556[1]. Please kindly review > and share > > >>>>> your feedback. > > >>>>> > > >>>>> > > >>>>> Thanks, > > >>>>> Xuefu > > >>>>> > > >>>>> [1] https://issues.apache.org/jira/browse/FLINK-10556 > > >>>>> ------------------------------------------------------------------ > > >>>>> Sender:Xuefu <[hidden email]> > > >>>>> Sent at:2018 Oct 25 (Thu) 01:08 > > >>>>> Recipient:Xuefu <[hidden email]>; Shuyi Chen < > > >>>>> [hidden email]> > > >>>>> Cc:yanghua1127 <[hidden email]>; Fabian Hueske < > > [hidden email]>; > > >>>>> dev <[hidden email]>; user <[hidden email]> > > >>>>> Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem > > >>>>> > > >>>>> Hi all, > > >>>>> > > >>>>> To wrap up the discussion, I have attached a PDF describing the > > > > >>>>> proposal, which is also attached to FLINK-10556 [1]. Please feel > free to > > >>>>> watch that JIRA to track the progress. > > >>>>> > > >>>>> Please also let me know if you have additional comments or > questions. > > >>>>> > > >>>>> Thanks, > > >>>>> Xuefu > > >>>>> > > >>>>> [1] https://issues.apache.org/jira/browse/FLINK-10556 > > >>>>> > > >>>>> > > >>>>> ------------------------------------------------------------------ > > >>>>> Sender:Xuefu <[hidden email]> > > >>>>> Sent at:2018 Oct 16 (Tue) 03:40 > > >>>>> Recipient:Shuyi Chen <[hidden email]> > > >>>>> Cc:yanghua1127 <[hidden email]>; Fabian Hueske < > > [hidden email]>; > > >>>>> dev <[hidden email]>; user <[hidden email]> > > >>>>> Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem > > >>>>> > > >>>>> Hi Shuyi, > > >>>>> > > > > >>>>> Thank you for your input. Yes, I agreed with a phased approach > and like > > > > >>>>> to move forward fast. :) We did some work internally on DDL > utilizing babel > > >>>>> parser in Calcite. While babel makes Calcite's grammar > extensible, at > > >>>>> first impression it still seems too cumbersome for a project when > too > > > > >>>>> much extensions are made. It's even challenging to find where the > extension > > > > >>>>> is needed! It would be certainly better if Calcite can magically > support > > > > >>>>> Hive QL by just turning on a flag, such as that for MYSQL_5. I > can also > > > > >>>>> see that this could mean a lot of work on Calcite. Nevertheless, > I will > > > > >>>>> bring up the discussion over there and to see what their > community thinks. > > >>>>> > > >>>>> Would mind to share more info about the proposal on DDL that you > > >>>>> mentioned? We can certainly collaborate on this. > > >>>>> > > >>>>> Thanks, > > >>>>> Xuefu > > >>>>> > > >>>>> ------------------------------------------------------------------ > > >>>>> Sender:Shuyi Chen <[hidden email]> > > >>>>> Sent at:2018 Oct 14 (Sun) 08:30 > > >>>>> Recipient:Xuefu <[hidden email]> > > >>>>> Cc:yanghua1127 <[hidden email]>; Fabian Hueske < > > [hidden email]>; > > >>>>> dev <[hidden email]>; user <[hidden email]> > > >>>>> Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem > > >>>>> > > >>>>> Welcome to the community and thanks for the great proposal, > Xuefu! I > > > > >>>>> think the proposal can be divided into 2 stages: making Flink to > support > > > > >>>>> Hive features, and make Hive to work with Flink. I agreed with > Timo that on > > > > >>>>> starting with a smaller scope, so we can make progress faster. As > for [6], > > > > >>>>> a proposal for DDL is already in progress, and will come after > the unified > > > > >>>>> SQL connector API is done. For supporting Hive syntax, we might > need to > > >>>>> work with the Calcite community, and a recent effort called babel > ( > > >>>>> https://issues.apache.org/jira/browse/CALCITE-2280) in Calcite > might > > >>>>> help here. > > >>>>> > > >>>>> Thanks > > >>>>> Shuyi > > >>>>> > > >>>>> On Wed, Oct 10, 2018 at 8:02 PM Zhang, Xuefu < > > [hidden email]> > > >>>>> wrote: > > >>>>> Hi Fabian/Vno, > > >>>>> > > > > >>>>> Thank you very much for your encouragement inquiry. Sorry that I > didn't > > > > >>>>> see Fabian's email until I read Vino's response just now. > (Somehow Fabian's > > >>>>> went to the spam folder.) > > >>>>> > > > > >>>>> My proposal contains long-term and short-terms goals. > Nevertheless, the > > >>>>> effort will focus on the following areas, including Fabian's list: > > >>>>> > > >>>>> 1. Hive metastore connectivity - This covers both read/write > access, > > > > >>>>> which means Flink can make full use of Hive's metastore as its > catalog (at > > >>>>> least for the batch but can extend for streaming as well). > > > > >>>>> 2. Metadata compatibility - Objects (databases, tables, > partitions, etc) > > > > >>>>> created by Hive can be understood by Flink and the reverse > direction is > > >>>>> true also. > > >>>>> 3. Data compatibility - Similar to #2, data produced by Hive can > be > > >>>>> consumed by Flink and vise versa. > > > > >>>>> 4. Support Hive UDFs - For all Hive's native udfs, Flink either > provides > > >>>>> its own implementation or make Hive's implementation work in > Flink. > > >>>>> Further, for user created UDFs in Hive, Flink SQL should provide a > > > > >>>>> mechanism allowing user to import them into Flink without any > code change > > >>>>> required. > > >>>>> 5. Data types - Flink SQL should support all data types that are > > >>>>> available in Hive. > > >>>>> 6. SQL Language - Flink SQL should support SQL standard (such as > > > > >>>>> SQL2003) with extension to support Hive's syntax and language > features, > > >>>>> around DDL, DML, and SELECT queries. > > > > >>>>> 7. SQL CLI - this is currently developing in Flink but more > effort is > > >>>>> needed. > > > > >>>>> 8. Server - provide a server that's compatible with Hive's > HiverServer2 > > > > >>>>> in thrift APIs, such that HiveServer2 users can reuse their > existing client > > >>>>> (such as beeline) but connect to Flink's thrift server instead. > > > > >>>>> 9. JDBC/ODBC drivers - Flink may provide its own JDBC/ODBC > drivers for > > >>>>> other application to use to connect to its thrift server > > >>>>> 10. Support other user's customizations in Hive, such as Hive > Serdes, > > >>>>> storage handlers, etc. > > > > >>>>> 11. Better task failure tolerance and task scheduling at Flink > runtime. > > >>>>> > > >>>>> As you can see, achieving all those requires significant effort > and > > > > >>>>> across all layers in Flink. However, a short-term goal could > include only > > > > >>>>> core areas (such as 1, 2, 4, 5, 6, 7) or start at a smaller > scope (such as > > >>>>> #3, #6). > > >>>>> > > > > >>>>> Please share your further thoughts. If we generally agree that > this is > > > > >>>>> the right direction, I could come up with a formal proposal > quickly and > > >>>>> then we can follow up with broader discussions. > > >>>>> > > >>>>> Thanks, > > >>>>> Xuefu > > >>>>> > > >>>>> > > >>>>> > > >>>>> ------------------------------------------------------------------ > > >>>>> Sender:vino yang <[hidden email]> > > >>>>> Sent at:2018 Oct 11 (Thu) 09:45 > > >>>>> Recipient:Fabian Hueske <[hidden email]> > > >>>>> Cc:dev <[hidden email]>; Xuefu <[hidden email] > > >; user < > > >>>>> [hidden email]> > > >>>>> Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem > > >>>>> > > >>>>> Hi Xuefu, > > >>>>> > > > > >>>>> Appreciate this proposal, and like Fabian, it would look better > if you > > >>>>> can give more details of the plan. > > >>>>> > > >>>>> Thanks, vino. > > >>>>> > > >>>>> Fabian Hueske <[hidden email]> 于2018年10月10日周三 下午5:27写道： > > >>>>> Hi Xuefu, > > >>>>> > > > > >>>>> Welcome to the Flink community and thanks for starting this > discussion! > > >>>>> Better Hive integration would be really great! > > >>>>> Can you go into details of what you are proposing? I can think of > a > > >>>>> couple ways to improve Flink in that regard: > > >>>>> > > >>>>> * Support for Hive UDFs > > >>>>> * Support for Hive metadata catalog > > >>>>> * Support for HiveQL syntax > > >>>>> * ??? > > >>>>> > > >>>>> Best, Fabian > > >>>>> > > >>>>> Am Di., 9. Okt. 2018 um 19:22 Uhr schrieb Zhang, Xuefu < > > >>>>> [hidden email]>: > > >>>>> Hi all, > > >>>>> > > >>>>> Along with the community's effort, inside Alibaba we have explored > > > > >>>>> Flink's potential as an execution engine not just for stream > processing but > > >>>>> also for batch processing. We are encouraged by our findings and > have > > > > >>>>> initiated our effort to make Flink's SQL capabilities > full-fledged. When > > > > >>>>> comparing what's available in Flink to the offerings from > competitive data > > > > >>>>> processing engines, we identified a major gap in Flink: a well > integration > > > > >>>>> with Hive ecosystem. This is crucial to the success of Flink SQL > and batch > > > > >>>>> due to the well-established data ecosystem around Hive. > Therefore, we have > > > > >>>>> done some initial work along this direction but there are still a > lot of > > >>>>> effort needed. > > >>>>> > > >>>>> We have two strategies in mind. The first one is to make Flink SQL > > > > >>>>> full-fledged and well-integrated with Hive ecosystem. This is a > similar > > > > >>>>> approach to what Spark SQL adopted. The second strategy is to > make Hive > > > > >>>>> itself work with Flink, similar to the proposal in [1]. Each > approach bears > > > > >>>>> its pros and cons, but they don’t need to be mutually exclusive > with each > > >>>>> targeting at different users and use cases. We believe that both > will > > >>>>> promote a much greater adoption of Flink beyond stream processing. > > >>>>> > > >>>>> We have been focused on the first approach and would like to > showcase > > > > >>>>> Flink's batch and SQL capabilities with Flink SQL. However, we > have also > > >>>>> planned to start strategy #2 as the follow-up effort. > > >>>>> > > > > >>>>> I'm completely new to Flink(, with a short bio [2] below), though > many > > > > >>>>> of my colleagues here at Alibaba are long-time contributors. > Nevertheless, > > > > >>>>> I'd like to share our thoughts and invite your early feedback. At > the same > > > > >>>>> time, I am working on a detailed proposal on Flink SQL's > integration with > > >>>>> Hive ecosystem, which will be also shared when ready. > > >>>>> > > >>>>> While the ideas are simple, each approach will demand significant > > > > >>>>> effort, more than what we can afford. Thus, the input and > contributions > > >>>>> from the communities are greatly welcome and appreciated. > > >>>>> > > >>>>> Regards, > > >>>>> > > >>>>> > > >>>>> Xuefu > > >>>>> > > >>>>> References: > > >>>>> > > >>>>> [1] https://issues.apache.org/jira/browse/HIVE-10712 > > > > >>>>> [2] Xuefu Zhang is a long-time open source veteran, worked or > working on > > >>>>> many projects under Apache Foundation, of which he is also an > honored > > > > >>>>> member. About 10 years ago he worked in the Hadoop team at Yahoo > where the > > > > >>>>> projects just got started. Later he worked at Cloudera, > initiating and > > > > >>>>> leading the development of Hive on Spark project in the > communities and > > > > >>>>> across many organizations. Prior to joining Alibaba, he worked at > Uber > > > > >>>>> where he promoted Hive on Spark to all Uber's SQL on Hadoop > workload and > > >>>>> significantly improved Uber's cluster efficiency. > > >>>>> > > >>>>> > > >>>>> > > >>>>> > > >>>>> -- > > > > >>>>> "So you have to trust that the dots will somehow connect in your > future." > > >>>>> > > >>>>> > > >>>>> -- > > > > >>>>> "So you have to trust that the dots will somehow connect in your > future." > > >>>>> > > > > > > |
Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem
|
|
Hi Eron,
Happy New Year! Thank you very much for your contribution, especially during the holidays. Wile I'm encouraged by your work, I'd also like to share my thoughts on how to move forward. First, please note that the design discussion is still finalizing, and we expect some moderate changes, especially around TableFactories. Another pending change is our decision to shy away from scala, which our work will be impacted by. Secondly, while your work seemed about plugging in catalogs definitions to the execution environment, which is less impacted by TableFactory change, I did notice some duplication of your work and ours. This is no big deal, but going forward, we should probable have a better communication on the work assignment so as to avoid any possible duplication of work. On the other hand, I think some of your work is interesting and valuable for inclusion once we finalize the overall design. Thus, please continue your research and experiment and let us know when you start working on anything so we can better coordinate. Thanks again for your interest and contributions. Thanks, Xuefu ------------------------------------------------------------------ From:Eron Wright <[hidden email]> Sent At:2019 Jan. 1 (Tue.) 18:39 To:dev <[hidden email]>; Xuefu <[hidden email]> Cc:Xiaowei Jiang <[hidden email]>; twalthr <[hidden email]>; piotr <[hidden email]>; Fabian Hueske <[hidden email]>; suez1224 <[hidden email]>; Bowen Li <[hidden email]> Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem Hi folks, there's clearly some incremental steps to be taken to introduce catalog support to SQL Client, complementary to what is proposed in the Flink-Hive Metastore design doc. I was quietly working on this over the holidays. I posted some new sub-tasks, PRs, and sample code to FLINK-10744. What inspired me to get involved is that the catalog interface seems like a great way to encapsulate a 'library' of Flink tables and functions. For example, the NYC Taxi dataset (TaxiRides, TaxiFares, various UDFs) may be nicely encapsulated as a catalog (TaxiData). Such a library should be fully consumable in SQL Client. I implemented the above. Some highlights: 1. A fully-worked example of using the Taxi dataset in SQL Client via an environment file. - an ASCII video showing the SQL Client in action: https://asciinema.org/a/C8xuAjmZSxCuApgFgZQyeIHuo - the corresponding environment file (will be even more concise once 'FLINK-10696 Catalog UDFs' is merged): https://github.com/EronWright/flink-training-exercises/blob/3be008d64be975ced0f1a7e3901a8c5353f72a7e/src/main/dist/conf/sql-client-defaults.yaml - the typed API for standalone table applications: https://github.com/EronWright/flink-training-exercises/blob/3be008d64be975ced0f1a7e3901a8c5353f72a7e/src/main/java/com/dataartisans/flinktraining/examples/table_java/examples/ViaCatalog.java#L50 2. Implementation of the core catalog descriptor and factory. I realize that some renames may later occur as per the design doc, and would be happy to do that as a follow-up. https://github.com/apache/flink/pull/7390 3. Implementation of a connect-style API on TableEnvironment to use catalog descriptor. https://github.com/apache/flink/pull/7392 4. Integration into SQL-Client's environment file: https://github.com/apache/flink/pull/7393 I realize that the overall Hive integration is still evolving, but I believe that these PRs are a good stepping stone. Here's the list (in bottom-up order): - https://github.com/apache/flink/pull/7386 - https://github.com/apache/flink/pull/7388 - https://github.com/apache/flink/pull/7389 - https://github.com/apache/flink/pull/7390 - https://github.com/apache/flink/pull/7392 - https://github.com/apache/flink/pull/7393 Thanks and enjoy 2019! Eron W On Sun, Nov 18, 2018 at 3:04 PM Zhang, Xuefu <[hidden email]> wrote: Hi Xiaowei, Thanks for bringing up the question. In the current design, the properties for meta objects are meant to cover anything that's specific to a particular catalog and agnostic to Flink. Anything that is common (such as schema for tables, query text for views, and udf classname) are abstracted as members of the respective classes. However, this is still in discussion, and Timo and I will go over this and provide an update. Please note that UDF is a little more involved than what the current design doc shows. I'm still refining this part. Thanks, Xuefu ------------------------------------------------------------------ Sender:Xiaowei Jiang <[hidden email]> Sent at:2018 Nov 18 (Sun) 15:17 Recipient:dev <[hidden email]> Cc:Xuefu <[hidden email]>; twalthr <[hidden email]>; piotr <[hidden email]>; Fabian Hueske <[hidden email]>; suez1224 <[hidden email]> Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem Thanks Xuefu for the detailed design doc! One question on the properties associated with the catalog objects. Are we going to leave them completely free form or we are going to set some standard for that? I think that the answer may depend on if we want to explore catalog specific optimization opportunities. In any case, I think that it might be helpful for standardize as much as possible into strongly typed classes and use leave these properties for catalog specific things. But I think that we can do it in steps. Xiaowei On Fri, Nov 16, 2018 at 4:00 AM Bowen Li <[hidden email]> wrote: Thanks for keeping on improving the overall design, Xuefu! It looks quite good to me now. Would be nice that cc-ed Flink committers can help to review and confirm! One minor suggestion: Since the last section of design doc already touches some new sql statements, shall we add another section in our doc and formalize the new sql statements in SQL Client and TableEnvironment that are gonna come along naturally with our design? Here are some that the design doc mentioned and some that I came up with: To be added: - USE <catalog> - set default catalog - USE <catalog.schema> - set default schema - SHOW CATALOGS - show all registered catalogs - SHOW SCHEMAS [FROM catalog] - list schemas in the current default catalog or the specified catalog - DESCRIBE VIEW view - show the view's definition in CatalogView - SHOW VIEWS [FROM schema/catalog.schema] - show views from current or a specified schema. (DDLs that can be addressed by either our design or Shuyi's DDL design) - CREATE/DROP/ALTER SCHEMA schema - CREATE/DROP/ALTER CATALOG catalog To be modified: - SHOW TABLES [FROM schema/catalog.schema] - show tables from current or a specified schema. Add 'from schema' to existing 'SHOW TABLES' statement - SHOW FUNCTIONS [FROM schema/catalog.schema] - show functions from current or a specified schema. Add 'from schema' to existing 'SHOW TABLES' statement' Thanks, Bowen On Wed, Nov 14, 2018 at 10:39 PM Zhang, Xuefu <[hidden email]> wrote: > Thanks, Bowen, for catching the error. I have granted comment permission > with the link. > > I also updated the doc with the latest class definitions. Everyone is > encouraged to review and comment. > > Thanks, > Xuefu > > ------------------------------------------------------------------ > Sender:Bowen Li <[hidden email]> > Sent at:2018 Nov 14 (Wed) 06:44 > Recipient:Xuefu <[hidden email]> > Cc:piotr <[hidden email]>; dev <[hidden email]>; Shuyi > Chen <[hidden email]> > Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem > > Hi Xuefu, > > Currently the new design doc > <https://docs.google.com/document/d/1Y9it78yaUvbv4g572ZK_lZnZaAGjqwM_EhjdOv4yJtw/edit> > is on “view only" mode, and people cannot leave comments. Can you please > change it to "can comment" or "can edit" mode? > > Thanks, Bowen > > > On Mon, Nov 12, 2018 at 9:51 PM Zhang, Xuefu <[hidden email]> > wrote: > Hi Piotr > > I have extracted the API portion of the design and the google doc is here > <https://docs.google.com/document/d/1Y9it78yaUvbv4g572ZK_lZnZaAGjqwM_EhjdOv4yJtw/edit?usp=sharing>. > Please review and provide your feedback. > > Thanks, > Xuefu > > ------------------------------------------------------------------ > Sender:Xuefu <[hidden email]> > Sent at:2018 Nov 12 (Mon) 12:43 > Recipient:Piotr Nowojski <[hidden email]>; dev < > [hidden email]> > Cc:Bowen Li <[hidden email]>; Shuyi Chen <[hidden email]> > Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem > > Hi Piotr, > > That sounds good to me. Let's close all the open questions ((there are a > couple of them)) in the Google doc and I should be able to quickly split > it into the three proposals as you suggested. > > Thanks, > Xuefu > > ------------------------------------------------------------------ > Sender:Piotr Nowojski <[hidden email]> > Sent at:2018 Nov 9 (Fri) 22:46 > Recipient:dev <[hidden email]>; Xuefu <[hidden email]> > Cc:Bowen Li <[hidden email]>; Shuyi Chen <[hidden email]> > Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem > > Hi, > > > Yes, it seems like the best solution. Maybe someone else can also suggests if we can split it further? Maybe changes in the interface in one doc, reading from hive meta store another and final storing our meta informations in hive meta store? > > Piotrek > > > On 9 Nov 2018, at 01:44, Zhang, Xuefu <[hidden email]> wrote: > > > > Hi Piotr, > > > > That seems to be good idea! > > > > > Since the google doc for the design is currently under extensive review, I will leave it as it is for now. However, I'll convert it to two different FLIPs when the time comes. > > > > How does it sound to you? > > > > Thanks, > > Xuefu > > > > > > ------------------------------------------------------------------ > > Sender:Piotr Nowojski <[hidden email]> > > Sent at:2018 Nov 9 (Fri) 02:31 > > Recipient:dev <[hidden email]> > > Cc:Bowen Li <[hidden email]>; Xuefu <[hidden email] > >; Shuyi Chen <[hidden email]> > > Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem > > > > Hi, > > > > > Maybe we should split this topic (and the design doc) into couple of smaller ones, hopefully independent. The questions that you have asked Fabian have for example very little to do with reading metadata from Hive Meta Store? > > > > Piotrek > > > >> On 7 Nov 2018, at 14:27, Fabian Hueske <[hidden email]> wrote: > >> > >> Hi Xuefu and all, > >> > >> Thanks for sharing this design document! > > >> I'm very much in favor of restructuring / reworking the catalog handling in > >> Flink SQL as outlined in the document. > > >> Most changes described in the design document seem to be rather general and > >> not specifically related to the Hive integration. > >> > > >> IMO, there are some aspects, especially those at the boundary of Hive and > >> Flink, that need a bit more discussion. For example > >> > >> * What does it take to make Flink schema compatible with Hive schema? > >> * How will Flink tables (descriptors) be stored in HMS? > >> * How do both Hive catalogs differ? Could they be integrated into to a > >> single one? When to use which one? > > >> * What meta information is provided by HMS? What of this can be leveraged > >> by Flink? > >> > >> Thank you, > >> Fabian > >> > >> Am Fr., 2. Nov. 2018 um 00:31 Uhr schrieb Bowen Li <[hidden email] > >: > >> > >>> After taking a look at how other discussion threads work, I think it's > >>> actually fine just keep our discussion here. It's up to you, Xuefu. > >>> > >>> The google doc LGTM. I left some minor comments. > >>> > >>> On Thu, Nov 1, 2018 at 10:17 AM Bowen Li <[hidden email]> wrote: > >>> > >>>> Hi all, > >>>> > >>>> As Xuefu has published the design doc on google, I agree with Shuyi's > > >>>> suggestion that we probably should start a new email thread like "[DISCUSS] > > >>>> ... Hive integration design ..." on only dev mailing list for community > >>>> devs to review. The current thread sends to both dev and user list. > >>>> > > >>>> This email thread is more like validating the general idea and direction > > >>>> with the community, and it's been pretty long and crowded so far. Since > > >>>> everyone is pro for the idea, we can move forward with another thread to > >>>> discuss and finalize the design. > >>>> > >>>> Thanks, > >>>> Bowen > >>>> > >>>> On Wed, Oct 31, 2018 at 12:16 PM Zhang, Xuefu < > [hidden email]> > >>>> wrote: > >>>> > >>>>> Hi Shuiyi, > >>>>> > > >>>>> Good idea. Actually the PDF was converted from a google doc. Here is its > >>>>> link: > >>>>> > >>>>> > https://docs.google.com/document/d/1SkppRD_rE3uOKSN-LuZCqn4f7dz0zW5aa6T_hBZq5_o/edit?usp=sharing > >>>>> Once we reach an agreement, I can convert it to a FLIP. > >>>>> > >>>>> Thanks, > >>>>> Xuefu > >>>>> > >>>>> > >>>>> > >>>>> ------------------------------------------------------------------ > >>>>> Sender:Shuyi Chen <[hidden email]> > >>>>> Sent at:2018 Nov 1 (Thu) 02:47 > >>>>> Recipient:Xuefu <[hidden email]> > >>>>> Cc:vino yang <[hidden email]>; Fabian Hueske < > [hidden email]>; > >>>>> dev <[hidden email]>; user <[hidden email]> > >>>>> Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem > >>>>> > >>>>> Hi Xuefu, > >>>>> > > >>>>> Thanks a lot for driving this big effort. I would suggest convert your > > >>>>> proposal and design doc into a google doc, and share it on the dev mailing > > >>>>> list for the community to review and comment with title like "[DISCUSS] ... > > >>>>> Hive integration design ..." . Once approved, we can document it as a FLIP > > >>>>> (Flink Improvement Proposals), and use JIRAs to track the implementations. > >>>>> What do you think? > >>>>> > >>>>> Shuyi > >>>>> > >>>>> On Tue, Oct 30, 2018 at 11:32 AM Zhang, Xuefu < > [hidden email]> > >>>>> wrote: > >>>>> Hi all, > >>>>> > >>>>> I have also shared a design doc on Hive metastore integration that is > > >>>>> attached here and also to FLINK-10556[1]. Please kindly review and share > >>>>> your feedback. > >>>>> > >>>>> > >>>>> Thanks, > >>>>> Xuefu > >>>>> > >>>>> [1] https://issues.apache.org/jira/browse/FLINK-10556 > >>>>> ------------------------------------------------------------------ > >>>>> Sender:Xuefu <[hidden email]> > >>>>> Sent at:2018 Oct 25 (Thu) 01:08 > >>>>> Recipient:Xuefu <[hidden email]>; Shuyi Chen < > >>>>> [hidden email]> > >>>>> Cc:yanghua1127 <[hidden email]>; Fabian Hueske < > [hidden email]>; > >>>>> dev <[hidden email]>; user <[hidden email]> > >>>>> Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem > >>>>> > >>>>> Hi all, > >>>>> > >>>>> To wrap up the discussion, I have attached a PDF describing the > > >>>>> proposal, which is also attached to FLINK-10556 [1]. Please feel free to > >>>>> watch that JIRA to track the progress. > >>>>> > >>>>> Please also let me know if you have additional comments or questions. > >>>>> > >>>>> Thanks, > >>>>> Xuefu > >>>>> > >>>>> [1] https://issues.apache.org/jira/browse/FLINK-10556 > >>>>> > >>>>> > >>>>> ------------------------------------------------------------------ > >>>>> Sender:Xuefu <[hidden email]> > >>>>> Sent at:2018 Oct 16 (Tue) 03:40 > >>>>> Recipient:Shuyi Chen <[hidden email]> > >>>>> Cc:yanghua1127 <[hidden email]>; Fabian Hueske < > [hidden email]>; > >>>>> dev <[hidden email]>; user <[hidden email]> > >>>>> Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem > >>>>> > >>>>> Hi Shuyi, > >>>>> > > >>>>> Thank you for your input. Yes, I agreed with a phased approach and like > > >>>>> to move forward fast. :) We did some work internally on DDL utilizing babel > >>>>> parser in Calcite. While babel makes Calcite's grammar extensible, at > >>>>> first impression it still seems too cumbersome for a project when too > > >>>>> much extensions are made. It's even challenging to find where the extension > > >>>>> is needed! It would be certainly better if Calcite can magically support > > >>>>> Hive QL by just turning on a flag, such as that for MYSQL_5. I can also > > >>>>> see that this could mean a lot of work on Calcite. Nevertheless, I will > > >>>>> bring up the discussion over there and to see what their community thinks. > >>>>> > >>>>> Would mind to share more info about the proposal on DDL that you > >>>>> mentioned? We can certainly collaborate on this. > >>>>> > >>>>> Thanks, > >>>>> Xuefu > >>>>> > >>>>> ------------------------------------------------------------------ > >>>>> Sender:Shuyi Chen <[hidden email]> > >>>>> Sent at:2018 Oct 14 (Sun) 08:30 > >>>>> Recipient:Xuefu <[hidden email]> > >>>>> Cc:yanghua1127 <[hidden email]>; Fabian Hueske < > [hidden email]>; > >>>>> dev <[hidden email]>; user <[hidden email]> > >>>>> Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem > >>>>> > >>>>> Welcome to the community and thanks for the great proposal, Xuefu! I > > >>>>> think the proposal can be divided into 2 stages: making Flink to support > > >>>>> Hive features, and make Hive to work with Flink. I agreed with Timo that on > > >>>>> starting with a smaller scope, so we can make progress faster. As for [6], > > >>>>> a proposal for DDL is already in progress, and will come after the unified > > >>>>> SQL connector API is done. For supporting Hive syntax, we might need to > >>>>> work with the Calcite community, and a recent effort called babel ( > >>>>> https://issues.apache.org/jira/browse/CALCITE-2280) in Calcite might > >>>>> help here. > >>>>> > >>>>> Thanks > >>>>> Shuyi > >>>>> > >>>>> On Wed, Oct 10, 2018 at 8:02 PM Zhang, Xuefu < > [hidden email]> > >>>>> wrote: > >>>>> Hi Fabian/Vno, > >>>>> > > >>>>> Thank you very much for your encouragement inquiry. Sorry that I didn't > > >>>>> see Fabian's email until I read Vino's response just now. (Somehow Fabian's > >>>>> went to the spam folder.) > >>>>> > > >>>>> My proposal contains long-term and short-terms goals. Nevertheless, the > >>>>> effort will focus on the following areas, including Fabian's list: > >>>>> > >>>>> 1. Hive metastore connectivity - This covers both read/write access, > > >>>>> which means Flink can make full use of Hive's metastore as its catalog (at > >>>>> least for the batch but can extend for streaming as well). > > >>>>> 2. Metadata compatibility - Objects (databases, tables, partitions, etc) > > >>>>> created by Hive can be understood by Flink and the reverse direction is > >>>>> true also. > >>>>> 3. Data compatibility - Similar to #2, data produced by Hive can be > >>>>> consumed by Flink and vise versa. > > >>>>> 4. Support Hive UDFs - For all Hive's native udfs, Flink either provides > >>>>> its own implementation or make Hive's implementation work in Flink. > >>>>> Further, for user created UDFs in Hive, Flink SQL should provide a > > >>>>> mechanism allowing user to import them into Flink without any code change > >>>>> required. > >>>>> 5. Data types - Flink SQL should support all data types that are > >>>>> available in Hive. > >>>>> 6. SQL Language - Flink SQL should support SQL standard (such as > > >>>>> SQL2003) with extension to support Hive's syntax and language features, > >>>>> around DDL, DML, and SELECT queries. > > >>>>> 7. SQL CLI - this is currently developing in Flink but more effort is > >>>>> needed. > > >>>>> 8. Server - provide a server that's compatible with Hive's HiverServer2 > > >>>>> in thrift APIs, such that HiveServer2 users can reuse their existing client > >>>>> (such as beeline) but connect to Flink's thrift server instead. > > >>>>> 9. JDBC/ODBC drivers - Flink may provide its own JDBC/ODBC drivers for > >>>>> other application to use to connect to its thrift server > >>>>> 10. Support other user's customizations in Hive, such as Hive Serdes, > >>>>> storage handlers, etc. > > >>>>> 11. Better task failure tolerance and task scheduling at Flink runtime. > >>>>> > >>>>> As you can see, achieving all those requires significant effort and > > >>>>> across all layers in Flink. However, a short-term goal could include only > > >>>>> core areas (such as 1, 2, 4, 5, 6, 7) or start at a smaller scope (such as > >>>>> #3, #6). > >>>>> > > >>>>> Please share your further thoughts. If we generally agree that this is > > >>>>> the right direction, I could come up with a formal proposal quickly and > >>>>> then we can follow up with broader discussions. > >>>>> > >>>>> Thanks, > >>>>> Xuefu > >>>>> > >>>>> > >>>>> > >>>>> ------------------------------------------------------------------ > >>>>> Sender:vino yang <[hidden email]> > >>>>> Sent at:2018 Oct 11 (Thu) 09:45 > >>>>> Recipient:Fabian Hueske <[hidden email]> > >>>>> Cc:dev <[hidden email]>; Xuefu <[hidden email] > >; user < > >>>>> [hidden email]> > >>>>> Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem > >>>>> > >>>>> Hi Xuefu, > >>>>> > > >>>>> Appreciate this proposal, and like Fabian, it would look better if you > >>>>> can give more details of the plan. > >>>>> > >>>>> Thanks, vino. > >>>>> > >>>>> Fabian Hueske <[hidden email]> 于2018年10月10日周三 下午5:27写道： > >>>>> Hi Xuefu, > >>>>> > > >>>>> Welcome to the Flink community and thanks for starting this discussion! > >>>>> Better Hive integration would be really great! > >>>>> Can you go into details of what you are proposing? I can think of a > >>>>> couple ways to improve Flink in that regard: > >>>>> > >>>>> * Support for Hive UDFs > >>>>> * Support for Hive metadata catalog > >>>>> * Support for HiveQL syntax > >>>>> * ??? > >>>>> > >>>>> Best, Fabian > >>>>> > >>>>> Am Di., 9. Okt. 2018 um 19:22 Uhr schrieb Zhang, Xuefu < > >>>>> [hidden email]>: > >>>>> Hi all, > >>>>> > >>>>> Along with the community's effort, inside Alibaba we have explored > > >>>>> Flink's potential as an execution engine not just for stream processing but > >>>>> also for batch processing. We are encouraged by our findings and have > > >>>>> initiated our effort to make Flink's SQL capabilities full-fledged. When > > >>>>> comparing what's available in Flink to the offerings from competitive data > > >>>>> processing engines, we identified a major gap in Flink: a well integration > > >>>>> with Hive ecosystem. This is crucial to the success of Flink SQL and batch > > >>>>> due to the well-established data ecosystem around Hive. Therefore, we have > > >>>>> done some initial work along this direction but there are still a lot of > >>>>> effort needed. > >>>>> > >>>>> We have two strategies in mind. The first one is to make Flink SQL > > >>>>> full-fledged and well-integrated with Hive ecosystem. This is a similar > > >>>>> approach to what Spark SQL adopted. The second strategy is to make Hive > > >>>>> itself work with Flink, similar to the proposal in [1]. Each approach bears > > >>>>> its pros and cons, but they don’t need to be mutually exclusive with each > >>>>> targeting at different users and use cases. We believe that both will > >>>>> promote a much greater adoption of Flink beyond stream processing. > >>>>> > >>>>> We have been focused on the first approach and would like to showcase > > >>>>> Flink's batch and SQL capabilities with Flink SQL. However, we have also > >>>>> planned to start strategy #2 as the follow-up effort. > >>>>> > > >>>>> I'm completely new to Flink(, with a short bio [2] below), though many > > >>>>> of my colleagues here at Alibaba are long-time contributors. Nevertheless, > > >>>>> I'd like to share our thoughts and invite your early feedback. At the same > > >>>>> time, I am working on a detailed proposal on Flink SQL's integration with > >>>>> Hive ecosystem, which will be also shared when ready. > >>>>> > >>>>> While the ideas are simple, each approach will demand significant > > >>>>> effort, more than what we can afford. Thus, the input and contributions > >>>>> from the communities are greatly welcome and appreciated. > >>>>> > >>>>> Regards, > >>>>> > >>>>> > >>>>> Xuefu > >>>>> > >>>>> References: > >>>>> > >>>>> [1] https://issues.apache.org/jira/browse/HIVE-10712 > > >>>>> [2] Xuefu Zhang is a long-time open source veteran, worked or working on > >>>>> many projects under Apache Foundation, of which he is also an honored > > >>>>> member. About 10 years ago he worked in the Hadoop team at Yahoo where the > > >>>>> projects just got started. Later he worked at Cloudera, initiating and > > >>>>> leading the development of Hive on Spark project in the communities and > > >>>>> across many organizations. Prior to joining Alibaba, he worked at Uber > > >>>>> where he promoted Hive on Spark to all Uber's SQL on Hadoop workload and > >>>>> significantly improved Uber's cluster efficiency. > >>>>> > >>>>> > >>>>> > >>>>> > >>>>> -- > > >>>>> "So you have to trust that the dots will somehow connect in your future." > >>>>> > >>>>> > >>>>> -- > > >>>>> "So you have to trust that the dots will somehow connect in your future." > >>>>> > > |
Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem
|
|
I propose that the community review and merge the PRs that I posted, and
then evolve the design thru 1.8 and beyond. I think having a basic infrastructure in place now will accelerate the effort, do you agree? Thanks again! On Wed, Jan 2, 2019 at 11:20 AM Zhang, Xuefu <[hidden email]> wrote: > Hi Eron, > > Happy New Year! > > Thank you very much for your contribution, especially during the holidays. > Wile I'm encouraged by your work, I'd also like to share my thoughts on how > to move forward. > > First, please note that the design discussion is still finalizing, and we > expect some moderate changes, especially around TableFactories. Another > pending change is our decision to shy away from scala, which our work will > be impacted by. > > Secondly, while your work seemed about plugging in catalogs definitions to > the execution environment, which is less impacted by TableFactory change, I > did notice some duplication of your work and ours. This is no big deal, but > going forward, we should probable have a better communication on the work > assignment so as to avoid any possible duplication of work. On the other > hand, I think some of your work is interesting and valuable for inclusion > once we finalize the overall design. > > Thus, please continue your research and experiment and let us know when > you start working on anything so we can better coordinate. > > Thanks again for your interest and contributions. > > Thanks, > Xuefu > > > > ------------------------------------------------------------------ > From:Eron Wright <[hidden email]> > Sent At:2019 Jan. 1 (Tue.) 18:39 > To:dev <[hidden email]>; Xuefu <[hidden email]> > Cc:Xiaowei Jiang <[hidden email]>; twalthr <[hidden email]>; > piotr <[hidden email]>; Fabian Hueske <[hidden email]>; > suez1224 <[hidden email]>; Bowen Li <[hidden email]> > Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem > > Hi folks, there's clearly some incremental steps to be taken to introduce > catalog support to SQL Client, complementary to what is proposed in the > Flink-Hive Metastore design doc. I was quietly working on this over the > holidays. I posted some new sub-tasks, PRs, and sample code > to FLINK-10744. > > What inspired me to get involved is that the catalog interface seems like > a great way to encapsulate a 'library' of Flink tables and functions. For > example, the NYC Taxi dataset (TaxiRides, TaxiFares, various UDFs) may be > nicely encapsulated as a catalog (TaxiData). Such a library should be > fully consumable in SQL Client. > > I implemented the above. Some highlights: > > 1. A fully-worked example of using the Taxi dataset in SQL Client via an > environment file. > - an ASCII video showing the SQL Client in action: > https://asciinema.org/a/C8xuAjmZSxCuApgFgZQyeIHuo > > - the corresponding environment file (will be even more concise once > 'FLINK-10696 Catalog UDFs' is merged): > *https://github.com/EronWright/flink-training-exercises/blob/3be008d64be975ced0f1a7e3901a8c5353f72a7e/src/main/dist/conf/sql-client-defaults.yaml > <https://github.com/EronWright/flink-training-exercises/blob/3be008d64be975ced0f1a7e3901a8c5353f72a7e/src/main/dist/conf/sql-client-defaults.yaml>* > > - the typed API for standalone table applications: > *https://github.com/EronWright/flink-training-exercises/blob/3be008d64be975ced0f1a7e3901a8c5353f72a7e/src/main/java/com/dataartisans/flinktraining/examples/table_java/examples/ViaCatalog.java#L50 > <https://github.com/EronWright/flink-training-exercises/blob/3be008d64be975ced0f1a7e3901a8c5353f72a7e/src/main/java/com/dataartisans/flinktraining/examples/table_java/examples/ViaCatalog.java#L50>* > > 2. Implementation of the core catalog descriptor and factory. I realize > that some renames may later occur as per the design doc, and would be happy > to do that as a follow-up. > https://github.com/apache/flink/pull/7390 > > 3. Implementation of a connect-style API on TableEnvironment to use > catalog descriptor. > https://github.com/apache/flink/pull/7392 > > 4. Integration into SQL-Client's environment file: > https://github.com/apache/flink/pull/7393 > > I realize that the overall Hive integration is still evolving, but I > believe that these PRs are a good stepping stone. Here's the list (in > bottom-up order): > - https://github.com/apache/flink/pull/7386 > - https://github.com/apache/flink/pull/7388 > - https://github.com/apache/flink/pull/7389 > - https://github.com/apache/flink/pull/7390 > - https://github.com/apache/flink/pull/7392 > - https://github.com/apache/flink/pull/7393 > > Thanks and enjoy 2019! > Eron W > > > On Sun, Nov 18, 2018 at 3:04 PM Zhang, Xuefu <[hidden email]> > wrote: > Hi Xiaowei, > > Thanks for bringing up the question. In the current design, the properties > for meta objects are meant to cover anything that's specific to a > particular catalog and agnostic to Flink. Anything that is common (such as > schema for tables, query text for views, and udf classname) are abstracted > as members of the respective classes. However, this is still in discussion, > and Timo and I will go over this and provide an update. > > Please note that UDF is a little more involved than what the current > design doc shows. I'm still refining this part. > > Thanks, > Xuefu > > > ------------------------------------------------------------------ > Sender:Xiaowei Jiang <[hidden email]> > Sent at:2018 Nov 18 (Sun) 15:17 > Recipient:dev <[hidden email]> > Cc:Xuefu <[hidden email]>; twalthr <[hidden email]>; piotr < > [hidden email]>; Fabian Hueske <[hidden email]>; suez1224 < > [hidden email]> > Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem > > Thanks Xuefu for the detailed design doc! One question on the properties > associated with the catalog objects. Are we going to leave them completely > free form or we are going to set some standard for that? I think that the > answer may depend on if we want to explore catalog specific optimization > opportunities. In any case, I think that it might be helpful for > standardize as much as possible into strongly typed classes and use leave > these properties for catalog specific things. But I think that we can do it > in steps. > > Xiaowei > On Fri, Nov 16, 2018 at 4:00 AM Bowen Li <[hidden email]> wrote: > Thanks for keeping on improving the overall design, Xuefu! It looks quite > good to me now. > > Would be nice that cc-ed Flink committers can help to review and confirm! > > > > One minor suggestion: Since the last section of design doc already touches > some new sql statements, shall we add another section in our doc and > formalize the new sql statements in SQL Client and TableEnvironment that > are gonna come along naturally with our design? Here are some that the > design doc mentioned and some that I came up with: > > To be added: > > - USE <catalog> - set default catalog > - USE <catalog.schema> - set default schema > - SHOW CATALOGS - show all registered catalogs > - SHOW SCHEMAS [FROM catalog] - list schemas in the current default > catalog or the specified catalog > - DESCRIBE VIEW view - show the view's definition in CatalogView > - SHOW VIEWS [FROM schema/catalog.schema] - show views from current or > a > specified schema. > > (DDLs that can be addressed by either our design or Shuyi's DDL design) > > - CREATE/DROP/ALTER SCHEMA schema > - CREATE/DROP/ALTER CATALOG catalog > > To be modified: > > - SHOW TABLES [FROM schema/catalog.schema] - show tables from current > or > a specified schema. Add 'from schema' to existing 'SHOW TABLES' > statement > - SHOW FUNCTIONS [FROM schema/catalog.schema] - show functions from > current or a specified schema. Add 'from schema' to existing 'SHOW > TABLES' > statement' > > > Thanks, Bowen > > > > On Wed, Nov 14, 2018 at 10:39 PM Zhang, Xuefu <[hidden email]> > wrote: > > > Thanks, Bowen, for catching the error. I have granted comment permission > > with the link. > > > > I also updated the doc with the latest class definitions. Everyone is > > encouraged to review and comment. > > > > Thanks, > > Xuefu > > > > ------------------------------------------------------------------ > > Sender:Bowen Li <[hidden email]> > > Sent at:2018 Nov 14 (Wed) 06:44 > > Recipient:Xuefu <[hidden email]> > > Cc:piotr <[hidden email]>; dev <[hidden email]>; Shuyi > > Chen <[hidden email]> > > Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem > > > > Hi Xuefu, > > > > Currently the new design doc > > < > https://docs.google.com/document/d/1Y9it78yaUvbv4g572ZK_lZnZaAGjqwM_EhjdOv4yJtw/edit > > > > is on “view only" mode, and people cannot leave comments. Can you please > > change it to "can comment" or "can edit" mode? > > > > Thanks, Bowen > > > > > > On Mon, Nov 12, 2018 at 9:51 PM Zhang, Xuefu <[hidden email]> > > wrote: > > Hi Piotr > > > > I have extracted the API portion of the design and the google doc is > here > > < > https://docs.google.com/document/d/1Y9it78yaUvbv4g572ZK_lZnZaAGjqwM_EhjdOv4yJtw/edit?usp=sharing > >. > > Please review and provide your feedback. > > > > Thanks, > > Xuefu > > > > ------------------------------------------------------------------ > > Sender:Xuefu <[hidden email]> > > Sent at:2018 Nov 12 (Mon) 12:43 > > Recipient:Piotr Nowojski <[hidden email]>; dev < > > [hidden email]> > > Cc:Bowen Li <[hidden email]>; Shuyi Chen <[hidden email]> > > Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem > > > > Hi Piotr, > > > > That sounds good to me. Let's close all the open questions ((there are a > > couple of them)) in the Google doc and I should be able to quickly split > > it into the three proposals as you suggested. > > > > Thanks, > > Xuefu > > > > ------------------------------------------------------------------ > > Sender:Piotr Nowojski <[hidden email]> > > Sent at:2018 Nov 9 (Fri) 22:46 > > Recipient:dev <[hidden email]>; Xuefu <[hidden email]> > > Cc:Bowen Li <[hidden email]>; Shuyi Chen <[hidden email]> > > Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem > > > > Hi, > > > > > > Yes, it seems like the best solution. Maybe someone else can also > suggests if we can split it further? Maybe changes in the interface in one > doc, reading from hive meta store another and final storing our meta > informations in hive meta store? > > > > Piotrek > > > > > On 9 Nov 2018, at 01:44, Zhang, Xuefu <[hidden email]> > wrote: > > > > > > Hi Piotr, > > > > > > That seems to be good idea! > > > > > > > > Since the google doc for the design is currently under extensive > review, I will leave it as it is for now. However, I'll convert it to two > different FLIPs when the time comes. > > > > > > How does it sound to you? > > > > > > Thanks, > > > Xuefu > > > > > > > > > ------------------------------------------------------------------ > > > Sender:Piotr Nowojski <[hidden email]> > > > Sent at:2018 Nov 9 (Fri) 02:31 > > > Recipient:dev <[hidden email]> > > > Cc:Bowen Li <[hidden email]>; Xuefu <[hidden email] > > >; Shuyi Chen <[hidden email]> > > > Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem > > > > > > Hi, > > > > > > > > Maybe we should split this topic (and the design doc) into couple of > smaller ones, hopefully independent. The questions that you have asked > Fabian have for example very little to do with reading metadata from Hive > Meta Store? > > > > > > Piotrek > > > > > >> On 7 Nov 2018, at 14:27, Fabian Hueske <[hidden email]> wrote: > > >> > > >> Hi Xuefu and all, > > >> > > >> Thanks for sharing this design document! > > > > >> I'm very much in favor of restructuring / reworking the catalog > handling in > > >> Flink SQL as outlined in the document. > > > > >> Most changes described in the design document seem to be rather > general and > > >> not specifically related to the Hive integration. > > >> > > > > >> IMO, there are some aspects, especially those at the boundary of > Hive and > > >> Flink, that need a bit more discussion. For example > > >> > > >> * What does it take to make Flink schema compatible with Hive schema? > > >> * How will Flink tables (descriptors) be stored in HMS? > > >> * How do both Hive catalogs differ? Could they be integrated into to > a > > >> single one? When to use which one? > > > > >> * What meta information is provided by HMS? What of this can be > leveraged > > >> by Flink? > > >> > > >> Thank you, > > >> Fabian > > >> > > >> Am Fr., 2. Nov. 2018 um 00:31 Uhr schrieb Bowen Li < > [hidden email] > > >: > > >> > > >>> After taking a look at how other discussion threads work, I think > it's > > >>> actually fine just keep our discussion here. It's up to you, Xuefu. > > >>> > > >>> The google doc LGTM. I left some minor comments. > > >>> > > >>> On Thu, Nov 1, 2018 at 10:17 AM Bowen Li <[hidden email]> > wrote: > > >>> > > >>>> Hi all, > > >>>> > > >>>> As Xuefu has published the design doc on google, I agree with > Shuyi's > > > > >>>> suggestion that we probably should start a new email thread like > "[DISCUSS] > > > > >>>> ... Hive integration design ..." on only dev mailing list for > community > > >>>> devs to review. The current thread sends to both dev and user list. > > >>>> > > > > >>>> This email thread is more like validating the general idea and > direction > > > > >>>> with the community, and it's been pretty long and crowded so far. > Since > > > > >>>> everyone is pro for the idea, we can move forward with another > thread to > > >>>> discuss and finalize the design. > > >>>> > > >>>> Thanks, > > >>>> Bowen > > >>>> > > >>>> On Wed, Oct 31, 2018 at 12:16 PM Zhang, Xuefu < > > [hidden email]> > > >>>> wrote: > > >>>> > > >>>>> Hi Shuiyi, > > >>>>> > > > > >>>>> Good idea. Actually the PDF was converted from a google doc. Here > is its > > >>>>> link: > > >>>>> > > >>>>> > > > https://docs.google.com/document/d/1SkppRD_rE3uOKSN-LuZCqn4f7dz0zW5aa6T_hBZq5_o/edit?usp=sharing > > >>>>> Once we reach an agreement, I can convert it to a FLIP. > > >>>>> > > >>>>> Thanks, > > >>>>> Xuefu > > >>>>> > > >>>>> > > >>>>> > > >>>>> ------------------------------------------------------------------ > > >>>>> Sender:Shuyi Chen <[hidden email]> > > >>>>> Sent at:2018 Nov 1 (Thu) 02:47 > > >>>>> Recipient:Xuefu <[hidden email]> > > >>>>> Cc:vino yang <[hidden email]>; Fabian Hueske < > > [hidden email]>; > > >>>>> dev <[hidden email]>; user <[hidden email]> > > >>>>> Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem > > >>>>> > > >>>>> Hi Xuefu, > > >>>>> > > > > >>>>> Thanks a lot for driving this big effort. I would suggest convert > your > > > > >>>>> proposal and design doc into a google doc, and share it on the > dev mailing > > > > >>>>> list for the community to review and comment with title like > "[DISCUSS] ... > > > > >>>>> Hive integration design ..." . Once approved, we can document it > as a FLIP > > > > >>>>> (Flink Improvement Proposals), and use JIRAs to track the > implementations. > > >>>>> What do you think? > > >>>>> > > >>>>> Shuyi > > >>>>> > > >>>>> On Tue, Oct 30, 2018 at 11:32 AM Zhang, Xuefu < > > [hidden email]> > > >>>>> wrote: > > >>>>> Hi all, > > >>>>> > > >>>>> I have also shared a design doc on Hive metastore integration > that is > > > > >>>>> attached here and also to FLINK-10556[1]. Please kindly review > and share > > >>>>> your feedback. > > >>>>> > > >>>>> > > >>>>> Thanks, > > >>>>> Xuefu > > >>>>> > > >>>>> [1] https://issues.apache.org/jira/browse/FLINK-10556 > > >>>>> ------------------------------------------------------------------ > > >>>>> Sender:Xuefu <[hidden email]> > > >>>>> Sent at:2018 Oct 25 (Thu) 01:08 > > >>>>> Recipient:Xuefu <[hidden email]>; Shuyi Chen < > > >>>>> [hidden email]> > > >>>>> Cc:yanghua1127 <[hidden email]>; Fabian Hueske < > > [hidden email]>; > > >>>>> dev <[hidden email]>; user <[hidden email]> > > >>>>> Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem > > >>>>> > > >>>>> Hi all, > > >>>>> > > >>>>> To wrap up the discussion, I have attached a PDF describing the > > > > >>>>> proposal, which is also attached to FLINK-10556 [1]. Please feel > free to > > >>>>> watch that JIRA to track the progress. > > >>>>> > > >>>>> Please also let me know if you have additional comments or > questions. > > >>>>> > > >>>>> Thanks, > > >>>>> Xuefu > > >>>>> > > >>>>> [1] https://issues.apache.org/jira/browse/FLINK-10556 > > >>>>> > > >>>>> > > >>>>> ------------------------------------------------------------------ > > >>>>> Sender:Xuefu <[hidden email]> > > >>>>> Sent at:2018 Oct 16 (Tue) 03:40 > > >>>>> Recipient:Shuyi Chen <[hidden email]> > > >>>>> Cc:yanghua1127 <[hidden email]>; Fabian Hueske < > > [hidden email]>; > > >>>>> dev <[hidden email]>; user <[hidden email]> > > >>>>> Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem > > >>>>> > > >>>>> Hi Shuyi, > > >>>>> > > > > >>>>> Thank you for your input. Yes, I agreed with a phased approach > and like > > > > >>>>> to move forward fast. :) We did some work internally on DDL > utilizing babel > > >>>>> parser in Calcite. While babel makes Calcite's grammar > extensible, at > > >>>>> first impression it still seems too cumbersome for a project when > too > > > > >>>>> much extensions are made. It's even challenging to find where the > extension > > > > >>>>> is needed! It would be certainly better if Calcite can magically > support > > > > >>>>> Hive QL by just turning on a flag, such as that for MYSQL_5. I > can also > > > > >>>>> see that this could mean a lot of work on Calcite. Nevertheless, > I will > > > > >>>>> bring up the discussion over there and to see what their > community thinks. > > >>>>> > > >>>>> Would mind to share more info about the proposal on DDL that you > > >>>>> mentioned? We can certainly collaborate on this. > > >>>>> > > >>>>> Thanks, > > >>>>> Xuefu > > >>>>> > > >>>>> ------------------------------------------------------------------ > > >>>>> Sender:Shuyi Chen <[hidden email]> > > >>>>> Sent at:2018 Oct 14 (Sun) 08:30 > > >>>>> Recipient:Xuefu <[hidden email]> > > >>>>> Cc:yanghua1127 <[hidden email]>; Fabian Hueske < > > [hidden email]>; > > >>>>> dev <[hidden email]>; user <[hidden email]> > > >>>>> Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem > > >>>>> > > >>>>> Welcome to the community and thanks for the great proposal, > Xuefu! I > > > > >>>>> think the proposal can be divided into 2 stages: making Flink to > support > > > > >>>>> Hive features, and make Hive to work with Flink. I agreed with > Timo that on > > > > >>>>> starting with a smaller scope, so we can make progress faster. As > for [6], > > > > >>>>> a proposal for DDL is already in progress, and will come after > the unified > > > > >>>>> SQL connector API is done. For supporting Hive syntax, we might > need to > > >>>>> work with the Calcite community, and a recent effort called babel > ( > > >>>>> https://issues.apache.org/jira/browse/CALCITE-2280) in Calcite > might > > >>>>> help here. > > >>>>> > > >>>>> Thanks > > >>>>> Shuyi > > >>>>> > > >>>>> On Wed, Oct 10, 2018 at 8:02 PM Zhang, Xuefu < > > [hidden email]> > > >>>>> wrote: > > >>>>> Hi Fabian/Vno, > > >>>>> > > > > >>>>> Thank you very much for your encouragement inquiry. Sorry that I > didn't > > > > >>>>> see Fabian's email until I read Vino's response just now. > (Somehow Fabian's > > >>>>> went to the spam folder.) > > >>>>> > > > > >>>>> My proposal contains long-term and short-terms goals. > Nevertheless, the > > >>>>> effort will focus on the following areas, including Fabian's list: > > >>>>> > > >>>>> 1. Hive metastore connectivity - This covers both read/write > access, > > > > >>>>> which means Flink can make full use of Hive's metastore as its > catalog (at > > >>>>> least for the batch but can extend for streaming as well). > > > > >>>>> 2. Metadata compatibility - Objects (databases, tables, > partitions, etc) > > > > >>>>> created by Hive can be understood by Flink and the reverse > direction is > > >>>>> true also. > > >>>>> 3. Data compatibility - Similar to #2, data produced by Hive can > be > > >>>>> consumed by Flink and vise versa. > > > > >>>>> 4. Support Hive UDFs - For all Hive's native udfs, Flink either > provides > > >>>>> its own implementation or make Hive's implementation work in > Flink. > > >>>>> Further, for user created UDFs in Hive, Flink SQL should provide a > > > > >>>>> mechanism allowing user to import them into Flink without any > code change > > >>>>> required. > > >>>>> 5. Data types - Flink SQL should support all data types that are > > >>>>> available in Hive. > > >>>>> 6. SQL Language - Flink SQL should support SQL standard (such as > > > > >>>>> SQL2003) with extension to support Hive's syntax and language > features, > > >>>>> around DDL, DML, and SELECT queries. > > > > >>>>> 7. SQL CLI - this is currently developing in Flink but more > effort is > > >>>>> needed. > > > > >>>>> 8. Server - provide a server that's compatible with Hive's > HiverServer2 > > > > >>>>> in thrift APIs, such that HiveServer2 users can reuse their > existing client > > >>>>> (such as beeline) but connect to Flink's thrift server instead. > > > > >>>>> 9. JDBC/ODBC drivers - Flink may provide its own JDBC/ODBC > drivers for > > >>>>> other application to use to connect to its thrift server > > >>>>> 10. Support other user's customizations in Hive, such as Hive > Serdes, > > >>>>> storage handlers, etc. > > > > >>>>> 11. Better task failure tolerance and task scheduling at Flink > runtime. > > >>>>> > > >>>>> As you can see, achieving all those requires significant effort > and > > > > >>>>> across all layers in Flink. However, a short-term goal could > include only > > > > >>>>> core areas (such as 1, 2, 4, 5, 6, 7) or start at a smaller > scope (such as > > >>>>> #3, #6). > > >>>>> > > > > >>>>> Please share your further thoughts. If we generally agree that > this is > > > > >>>>> the right direction, I could come up with a formal proposal > quickly and > > >>>>> then we can follow up with broader discussions. > > >>>>> > > >>>>> Thanks, > > >>>>> Xuefu > > >>>>> > > >>>>> > > >>>>> > > >>>>> ------------------------------------------------------------------ > > >>>>> Sender:vino yang <[hidden email]> > > >>>>> Sent at:2018 Oct 11 (Thu) 09:45 > > >>>>> Recipient:Fabian Hueske <[hidden email]> > > >>>>> Cc:dev <[hidden email]>; Xuefu <[hidden email] > > >; user < > > >>>>> [hidden email]> > > >>>>> Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem > > >>>>> > > >>>>> Hi Xuefu, > > >>>>> > > > > >>>>> Appreciate this proposal, and like Fabian, it would look better > if you > > >>>>> can give more details of the plan. > > >>>>> > > >>>>> Thanks, vino. > > >>>>> > > >>>>> Fabian Hueske <[hidden email]> 于2018年10月10日周三 下午5:27写道： > > >>>>> Hi Xuefu, > > >>>>> > > > > >>>>> Welcome to the Flink community and thanks for starting this > discussion! > > >>>>> Better Hive integration would be really great! > > >>>>> Can you go into details of what you are proposing? I can think of > a > > >>>>> couple ways to improve Flink in that regard: > > >>>>> > > >>>>> * Support for Hive UDFs > > >>>>> * Support for Hive metadata catalog > > >>>>> * Support for HiveQL syntax > > >>>>> * ??? > > >>>>> > > >>>>> Best, Fabian > > >>>>> > > >>>>> Am Di., 9. Okt. 2018 um 19:22 Uhr schrieb Zhang, Xuefu < > > >>>>> [hidden email]>: > > >>>>> Hi all, > > >>>>> > > >>>>> Along with the community's effort, inside Alibaba we have explored > > > > >>>>> Flink's potential as an execution engine not just for stream > processing but > > >>>>> also for batch processing. We are encouraged by our findings and > have > > > > >>>>> initiated our effort to make Flink's SQL capabilities > full-fledged. When > > > > >>>>> comparing what's available in Flink to the offerings from > competitive data > > > > >>>>> processing engines, we identified a major gap in Flink: a well > integration > > > > >>>>> with Hive ecosystem. This is crucial to the success of Flink SQL > and batch > > > > >>>>> due to the well-established data ecosystem around Hive. > Therefore, we have > > > > >>>>> done some initial work along this direction but there are still a > lot of > > >>>>> effort needed. > > >>>>> > > >>>>> We have two strategies in mind. The first one is to make Flink SQL > > > > >>>>> full-fledged and well-integrated with Hive ecosystem. This is a > similar > > > > >>>>> approach to what Spark SQL adopted. The second strategy is to > make Hive > > > > >>>>> itself work with Flink, similar to the proposal in [1]. Each > approach bears > > > > >>>>> its pros and cons, but they don’t need to be mutually exclusive > with each > > >>>>> targeting at different users and use cases. We believe that both > will > > >>>>> promote a much greater adoption of Flink beyond stream processing. > > >>>>> > > >>>>> We have been focused on the first approach and would like to > showcase > > > > >>>>> Flink's batch and SQL capabilities with Flink SQL. However, we > have also > > >>>>> planned to start strategy #2 as the follow-up effort. > > >>>>> > > > > >>>>> I'm completely new to Flink(, with a short bio [2] below), though > many > > > > >>>>> of my colleagues here at Alibaba are long-time contributors. > Nevertheless, > > > > >>>>> I'd like to share our thoughts and invite your early feedback. At > the same > > > > >>>>> time, I am working on a detailed proposal on Flink SQL's > integration with > > >>>>> Hive ecosystem, which will be also shared when ready. > > >>>>> > > >>>>> While the ideas are simple, each approach will demand significant > > > > >>>>> effort, more than what we can afford. Thus, the input and > contributions > > >>>>> from the communities are greatly welcome and appreciated. > > >>>>> > > >>>>> Regards, > > >>>>> > > >>>>> > > >>>>> Xuefu > > >>>>> > > >>>>> References: > > >>>>> > > >>>>> [1] https://issues.apache.org/jira/browse/HIVE-10712 > > > > >>>>> [2] Xuefu Zhang is a long-time open source veteran, worked or > working on > > >>>>> many projects under Apache Foundation, of which he is also an > honored > > > > >>>>> member. About 10 years ago he worked in the Hadoop team at Yahoo > where the > > > > >>>>> projects just got started. Later he worked at Cloudera, > initiating and > > > > >>>>> leading the development of Hive on Spark project in the > communities and > > > > >>>>> across many organizations. Prior to joining Alibaba, he worked at > Uber > > > > >>>>> where he promoted Hive on Spark to all Uber's SQL on Hadoop > workload and > > >>>>> significantly improved Uber's cluster efficiency. > > >>>>> > > >>>>> > > >>>>> > > >>>>> > > >>>>> -- > > > > >>>>> "So you have to trust that the dots will somehow connect in your > future." > > >>>>> > > >>>>> > > >>>>> -- > > > > >>>>> "So you have to trust that the dots will somehow connect in your > future." > > >>>>> > > > > > > |
Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem
|
|
Would a couple folks raise their hand to make a review pass thru the 6 PRs
listed above? It is a lovely stack of PRs that is 'all green' at the moment. I would be happy to open follow-on PRs to rapidly align with other efforts. Note that the code is agnostic to the details of the ExternalCatalog interface; the code would not be obsolete if/when the catalog interface is enhanced as per the design doc. On Wed, Jan 2, 2019 at 1:35 PM Eron Wright <[hidden email]> wrote: > I propose that the community review and merge the PRs that I posted, and > then evolve the design thru 1.8 and beyond. I think having a basic > infrastructure in place now will accelerate the effort, do you agree? > > Thanks again! > > On Wed, Jan 2, 2019 at 11:20 AM Zhang, Xuefu <[hidden email]> > wrote: > >> Hi Eron, >> >> Happy New Year! >> >> Thank you very much for your contribution, especially during the >> holidays. Wile I'm encouraged by your work, I'd also like to share my >> thoughts on how to move forward. >> >> First, please note that the design discussion is still finalizing, and we >> expect some moderate changes, especially around TableFactories. Another >> pending change is our decision to shy away from scala, which our work will >> be impacted by. >> >> Secondly, while your work seemed about plugging in catalogs definitions >> to the execution environment, which is less impacted by TableFactory >> change, I did notice some duplication of your work and ours. This is no big >> deal, but going forward, we should probable have a better communication on >> the work assignment so as to avoid any possible duplication of work. On the >> other hand, I think some of your work is interesting and valuable for >> inclusion once we finalize the overall design. >> >> Thus, please continue your research and experiment and let us know when >> you start working on anything so we can better coordinate. >> >> Thanks again for your interest and contributions. >> >> Thanks, >> Xuefu >> >> >> >> ------------------------------------------------------------------ >> From:Eron Wright <[hidden email]> >> Sent At:2019 Jan. 1 (Tue.) 18:39 >> To:dev <[hidden email]>; Xuefu <[hidden email]> >> Cc:Xiaowei Jiang <[hidden email]>; twalthr <[hidden email]>; >> piotr <[hidden email]>; Fabian Hueske <[hidden email]>; >> suez1224 <[hidden email]>; Bowen Li <[hidden email]> >> Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem >> >> Hi folks, there's clearly some incremental steps to be taken to introduce >> catalog support to SQL Client, complementary to what is proposed in the >> Flink-Hive Metastore design doc. I was quietly working on this over the >> holidays. I posted some new sub-tasks, PRs, and sample code >> to FLINK-10744. >> >> What inspired me to get involved is that the catalog interface seems like >> a great way to encapsulate a 'library' of Flink tables and functions. For >> example, the NYC Taxi dataset (TaxiRides, TaxiFares, various UDFs) may be >> nicely encapsulated as a catalog (TaxiData). Such a library should be >> fully consumable in SQL Client. >> >> I implemented the above. Some highlights: >> >> 1. A fully-worked example of using the Taxi dataset in SQL Client via an >> environment file. >> - an ASCII video showing the SQL Client in action: >> https://asciinema.org/a/C8xuAjmZSxCuApgFgZQyeIHuo >> >> - the corresponding environment file (will be even more concise once >> 'FLINK-10696 Catalog UDFs' is merged): >> *https://github.com/EronWright/flink-training-exercises/blob/3be008d64be975ced0f1a7e3901a8c5353f72a7e/src/main/dist/conf/sql-client-defaults.yaml >> <https://github.com/EronWright/flink-training-exercises/blob/3be008d64be975ced0f1a7e3901a8c5353f72a7e/src/main/dist/conf/sql-client-defaults.yaml>* >> >> - the typed API for standalone table applications: >> *https://github.com/EronWright/flink-training-exercises/blob/3be008d64be975ced0f1a7e3901a8c5353f72a7e/src/main/java/com/dataartisans/flinktraining/examples/table_java/examples/ViaCatalog.java#L50 >> <https://github.com/EronWright/flink-training-exercises/blob/3be008d64be975ced0f1a7e3901a8c5353f72a7e/src/main/java/com/dataartisans/flinktraining/examples/table_java/examples/ViaCatalog.java#L50>* >> >> 2. Implementation of the core catalog descriptor and factory. I realize >> that some renames may later occur as per the design doc, and would be happy >> to do that as a follow-up. >> https://github.com/apache/flink/pull/7390 >> >> 3. Implementation of a connect-style API on TableEnvironment to use >> catalog descriptor. >> https://github.com/apache/flink/pull/7392 >> >> 4. Integration into SQL-Client's environment file: >> https://github.com/apache/flink/pull/7393 >> >> I realize that the overall Hive integration is still evolving, but I >> believe that these PRs are a good stepping stone. Here's the list (in >> bottom-up order): >> - https://github.com/apache/flink/pull/7386 >> - https://github.com/apache/flink/pull/7388 >> - https://github.com/apache/flink/pull/7389 >> - https://github.com/apache/flink/pull/7390 >> - https://github.com/apache/flink/pull/7392 >> - https://github.com/apache/flink/pull/7393 >> >> Thanks and enjoy 2019! >> Eron W >> >> >> On Sun, Nov 18, 2018 at 3:04 PM Zhang, Xuefu <[hidden email]> >> wrote: >> Hi Xiaowei, >> >> Thanks for bringing up the question. In the current design, the >> properties for meta objects are meant to cover anything that's specific to >> a particular catalog and agnostic to Flink. Anything that is common (such >> as schema for tables, query text for views, and udf classname) are >> abstracted as members of the respective classes. However, this is still in >> discussion, and Timo and I will go over this and provide an update. >> >> Please note that UDF is a little more involved than what the current >> design doc shows. I'm still refining this part. >> >> Thanks, >> Xuefu >> >> >> ------------------------------------------------------------------ >> Sender:Xiaowei Jiang <[hidden email]> >> Sent at:2018 Nov 18 (Sun) 15:17 >> Recipient:dev <[hidden email]> >> Cc:Xuefu <[hidden email]>; twalthr <[hidden email]>; piotr < >> [hidden email]>; Fabian Hueske <[hidden email]>; suez1224 < >> [hidden email]> >> Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem >> >> Thanks Xuefu for the detailed design doc! One question on the properties >> associated with the catalog objects. Are we going to leave them completely >> free form or we are going to set some standard for that? I think that the >> answer may depend on if we want to explore catalog specific optimization >> opportunities. In any case, I think that it might be helpful for >> standardize as much as possible into strongly typed classes and use leave >> these properties for catalog specific things. But I think that we can do it >> in steps. >> >> Xiaowei >> On Fri, Nov 16, 2018 at 4:00 AM Bowen Li <[hidden email]> wrote: >> Thanks for keeping on improving the overall design, Xuefu! It looks quite >> good to me now. >> >> Would be nice that cc-ed Flink committers can help to review and confirm! >> >> >> >> One minor suggestion: Since the last section of design doc already >> touches >> some new sql statements, shall we add another section in our doc and >> formalize the new sql statements in SQL Client and TableEnvironment that >> are gonna come along naturally with our design? Here are some that the >> design doc mentioned and some that I came up with: >> >> To be added: >> >> - USE <catalog> - set default catalog >> - USE <catalog.schema> - set default schema >> - SHOW CATALOGS - show all registered catalogs >> - SHOW SCHEMAS [FROM catalog] - list schemas in the current default >> catalog or the specified catalog >> - DESCRIBE VIEW view - show the view's definition in CatalogView >> - SHOW VIEWS [FROM schema/catalog.schema] - show views from current >> or a >> specified schema. >> >> (DDLs that can be addressed by either our design or Shuyi's DDL >> design) >> >> - CREATE/DROP/ALTER SCHEMA schema >> - CREATE/DROP/ALTER CATALOG catalog >> >> To be modified: >> >> - SHOW TABLES [FROM schema/catalog.schema] - show tables from current >> or >> a specified schema. Add 'from schema' to existing 'SHOW TABLES' >> statement >> - SHOW FUNCTIONS [FROM schema/catalog.schema] - show functions from >> current or a specified schema. Add 'from schema' to existing 'SHOW >> TABLES' >> statement' >> >> >> Thanks, Bowen >> >> >> >> On Wed, Nov 14, 2018 at 10:39 PM Zhang, Xuefu <[hidden email]> >> wrote: >> >> > Thanks, Bowen, for catching the error. I have granted comment >> permission >> > with the link. >> > >> > I also updated the doc with the latest class definitions. Everyone is >> > encouraged to review and comment. >> > >> > Thanks, >> > Xuefu >> > >> > ------------------------------------------------------------------ >> > Sender:Bowen Li <[hidden email]> >> > Sent at:2018 Nov 14 (Wed) 06:44 >> > Recipient:Xuefu <[hidden email]> >> > Cc:piotr <[hidden email]>; dev <[hidden email]>; Shuyi >> > Chen <[hidden email]> >> > Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem >> > >> > Hi Xuefu, >> > >> > Currently the new design doc >> > < >> https://docs.google.com/document/d/1Y9it78yaUvbv4g572ZK_lZnZaAGjqwM_EhjdOv4yJtw/edit >> > >> > is on “view only" mode, and people cannot leave comments. Can you >> please >> > change it to "can comment" or "can edit" mode? >> > >> > Thanks, Bowen >> > >> > >> > On Mon, Nov 12, 2018 at 9:51 PM Zhang, Xuefu <[hidden email]> >> > wrote: >> > Hi Piotr >> > >> > I have extracted the API portion of the design and the google doc is >> here >> > < >> https://docs.google.com/document/d/1Y9it78yaUvbv4g572ZK_lZnZaAGjqwM_EhjdOv4yJtw/edit?usp=sharing >> >. >> > Please review and provide your feedback. >> > >> > Thanks, >> > Xuefu >> > >> > ------------------------------------------------------------------ >> > Sender:Xuefu <[hidden email]> >> > Sent at:2018 Nov 12 (Mon) 12:43 >> > Recipient:Piotr Nowojski <[hidden email]>; dev < >> > [hidden email]> >> > Cc:Bowen Li <[hidden email]>; Shuyi Chen <[hidden email]> >> > Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem >> > >> > Hi Piotr, >> > >> > That sounds good to me. Let's close all the open questions ((there are >> a >> > couple of them)) in the Google doc and I should be able to quickly >> split >> > it into the three proposals as you suggested. >> > >> > Thanks, >> > Xuefu >> > >> > ------------------------------------------------------------------ >> > Sender:Piotr Nowojski <[hidden email]> >> > Sent at:2018 Nov 9 (Fri) 22:46 >> > Recipient:dev <[hidden email]>; Xuefu <[hidden email]> >> > Cc:Bowen Li <[hidden email]>; Shuyi Chen <[hidden email]> >> > Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem >> > >> > Hi, >> > >> > >> > Yes, it seems like the best solution. Maybe someone else can also >> suggests if we can split it further? Maybe changes in the interface in one >> doc, reading from hive meta store another and final storing our meta >> informations in hive meta store? >> > >> > Piotrek >> > >> > > On 9 Nov 2018, at 01:44, Zhang, Xuefu <[hidden email]> >> wrote: >> > > >> > > Hi Piotr, >> > > >> > > That seems to be good idea! >> > > >> > >> > > Since the google doc for the design is currently under extensive >> review, I will leave it as it is for now. However, I'll convert it to two >> different FLIPs when the time comes. >> > > >> > > How does it sound to you? >> > > >> > > Thanks, >> > > Xuefu >> > > >> > > >> > > ------------------------------------------------------------------ >> > > Sender:Piotr Nowojski <[hidden email]> >> > > Sent at:2018 Nov 9 (Fri) 02:31 >> > > Recipient:dev <[hidden email]> >> > > Cc:Bowen Li <[hidden email]>; Xuefu <[hidden email] >> > >; Shuyi Chen <[hidden email]> >> > > Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive ecosystem >> > > >> > > Hi, >> > > >> > >> > > Maybe we should split this topic (and the design doc) into couple of >> smaller ones, hopefully independent. The questions that you have asked >> Fabian have for example very little to do with reading metadata from Hive >> Meta Store? >> > > >> > > Piotrek >> > > >> > >> On 7 Nov 2018, at 14:27, Fabian Hueske <[hidden email]> wrote: >> > >> >> > >> Hi Xuefu and all, >> > >> >> > >> Thanks for sharing this design document! >> > >> > >> I'm very much in favor of restructuring / reworking the catalog >> handling in >> > >> Flink SQL as outlined in the document. >> > >> > >> Most changes described in the design document seem to be rather >> general and >> > >> not specifically related to the Hive integration. >> > >> >> > >> > >> IMO, there are some aspects, especially those at the boundary of >> Hive and >> > >> Flink, that need a bit more discussion. For example >> > >> >> > >> * What does it take to make Flink schema compatible with Hive >> schema? >> > >> * How will Flink tables (descriptors) be stored in HMS? >> > >> * How do both Hive catalogs differ? Could they be integrated into >> to a >> > >> single one? When to use which one? >> > >> > >> * What meta information is provided by HMS? What of this can be >> leveraged >> > >> by Flink? >> > >> >> > >> Thank you, >> > >> Fabian >> > >> >> > >> Am Fr., 2. Nov. 2018 um 00:31 Uhr schrieb Bowen Li < >> [hidden email] >> > >: >> > >> >> > >>> After taking a look at how other discussion threads work, I think >> it's >> > >>> actually fine just keep our discussion here. It's up to you, Xuefu. >> > >>> >> > >>> The google doc LGTM. I left some minor comments. >> > >>> >> > >>> On Thu, Nov 1, 2018 at 10:17 AM Bowen Li <[hidden email]> >> wrote: >> > >>> >> > >>>> Hi all, >> > >>>> >> > >>>> As Xuefu has published the design doc on google, I agree with >> Shuyi's >> > >> > >>>> suggestion that we probably should start a new email thread like >> "[DISCUSS] >> > >> > >>>> ... Hive integration design ..." on only dev mailing list for >> community >> > >>>> devs to review. The current thread sends to both dev and user >> list. >> > >>>> >> > >> > >>>> This email thread is more like validating the general idea and >> direction >> > >> > >>>> with the community, and it's been pretty long and crowded so far. >> Since >> > >> > >>>> everyone is pro for the idea, we can move forward with another >> thread to >> > >>>> discuss and finalize the design. >> > >>>> >> > >>>> Thanks, >> > >>>> Bowen >> > >>>> >> > >>>> On Wed, Oct 31, 2018 at 12:16 PM Zhang, Xuefu < >> > [hidden email]> >> > >>>> wrote: >> > >>>> >> > >>>>> Hi Shuiyi, >> > >>>>> >> > >> > >>>>> Good idea. Actually the PDF was converted from a google doc. >> Here is its >> > >>>>> link: >> > >>>>> >> > >>>>> >> > >> https://docs.google.com/document/d/1SkppRD_rE3uOKSN-LuZCqn4f7dz0zW5aa6T_hBZq5_o/edit?usp=sharing >> > >>>>> Once we reach an agreement, I can convert it to a FLIP. >> > >>>>> >> > >>>>> Thanks, >> > >>>>> Xuefu >> > >>>>> >> > >>>>> >> > >>>>> >> > >>>>> >> ------------------------------------------------------------------ >> > >>>>> Sender:Shuyi Chen <[hidden email]> >> > >>>>> Sent at:2018 Nov 1 (Thu) 02:47 >> > >>>>> Recipient:Xuefu <[hidden email]> >> > >>>>> Cc:vino yang <[hidden email]>; Fabian Hueske < >> > [hidden email]>; >> > >>>>> dev <[hidden email]>; user <[hidden email]> >> > >>>>> Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive >> ecosystem >> > >>>>> >> > >>>>> Hi Xuefu, >> > >>>>> >> > >> > >>>>> Thanks a lot for driving this big effort. I would suggest >> convert your >> > >> > >>>>> proposal and design doc into a google doc, and share it on the >> dev mailing >> > >> > >>>>> list for the community to review and comment with title like >> "[DISCUSS] ... >> > >> > >>>>> Hive integration design ..." . Once approved, we can document >> it as a FLIP >> > >> > >>>>> (Flink Improvement Proposals), and use JIRAs to track the >> implementations. >> > >>>>> What do you think? >> > >>>>> >> > >>>>> Shuyi >> > >>>>> >> > >>>>> On Tue, Oct 30, 2018 at 11:32 AM Zhang, Xuefu < >> > [hidden email]> >> > >>>>> wrote: >> > >>>>> Hi all, >> > >>>>> >> > >>>>> I have also shared a design doc on Hive metastore integration >> that is >> > >> > >>>>> attached here and also to FLINK-10556[1]. Please kindly review >> and share >> > >>>>> your feedback. >> > >>>>> >> > >>>>> >> > >>>>> Thanks, >> > >>>>> Xuefu >> > >>>>> >> > >>>>> [1] https://issues.apache.org/jira/browse/FLINK-10556 >> > >>>>> >> ------------------------------------------------------------------ >> > >>>>> Sender:Xuefu <[hidden email]> >> > >>>>> Sent at:2018 Oct 25 (Thu) 01:08 >> > >>>>> Recipient:Xuefu <[hidden email]>; Shuyi Chen < >> > >>>>> [hidden email]> >> > >>>>> Cc:yanghua1127 <[hidden email]>; Fabian Hueske < >> > [hidden email]>; >> > >>>>> dev <[hidden email]>; user <[hidden email]> >> > >>>>> Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive >> ecosystem >> > >>>>> >> > >>>>> Hi all, >> > >>>>> >> > >>>>> To wrap up the discussion, I have attached a PDF describing the >> > >> > >>>>> proposal, which is also attached to FLINK-10556 [1]. Please feel >> free to >> > >>>>> watch that JIRA to track the progress. >> > >>>>> >> > >>>>> Please also let me know if you have additional comments or >> questions. >> > >>>>> >> > >>>>> Thanks, >> > >>>>> Xuefu >> > >>>>> >> > >>>>> [1] https://issues.apache.org/jira/browse/FLINK-10556 >> > >>>>> >> > >>>>> >> > >>>>> >> ------------------------------------------------------------------ >> > >>>>> Sender:Xuefu <[hidden email]> >> > >>>>> Sent at:2018 Oct 16 (Tue) 03:40 >> > >>>>> Recipient:Shuyi Chen <[hidden email]> >> > >>>>> Cc:yanghua1127 <[hidden email]>; Fabian Hueske < >> > [hidden email]>; >> > >>>>> dev <[hidden email]>; user <[hidden email]> >> > >>>>> Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive >> ecosystem >> > >>>>> >> > >>>>> Hi Shuyi, >> > >>>>> >> > >> > >>>>> Thank you for your input. Yes, I agreed with a phased approach >> and like >> > >> > >>>>> to move forward fast. :) We did some work internally on DDL >> utilizing babel >> > >>>>> parser in Calcite. While babel makes Calcite's grammar >> extensible, at >> > >>>>> first impression it still seems too cumbersome for a project >> when too >> > >> > >>>>> much extensions are made. It's even challenging to find where >> the extension >> > >> > >>>>> is needed! It would be certainly better if Calcite can magically >> support >> > >> > >>>>> Hive QL by just turning on a flag, such as that for MYSQL_5. I >> can also >> > >> > >>>>> see that this could mean a lot of work on Calcite. Nevertheless, >> I will >> > >> > >>>>> bring up the discussion over there and to see what their >> community thinks. >> > >>>>> >> > >>>>> Would mind to share more info about the proposal on DDL that you >> > >>>>> mentioned? We can certainly collaborate on this. >> > >>>>> >> > >>>>> Thanks, >> > >>>>> Xuefu >> > >>>>> >> > >>>>> >> ------------------------------------------------------------------ >> > >>>>> Sender:Shuyi Chen <[hidden email]> >> > >>>>> Sent at:2018 Oct 14 (Sun) 08:30 >> > >>>>> Recipient:Xuefu <[hidden email]> >> > >>>>> Cc:yanghua1127 <[hidden email]>; Fabian Hueske < >> > [hidden email]>; >> > >>>>> dev <[hidden email]>; user <[hidden email]> >> > >>>>> Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive >> ecosystem >> > >>>>> >> > >>>>> Welcome to the community and thanks for the great proposal, >> Xuefu! I >> > >> > >>>>> think the proposal can be divided into 2 stages: making Flink to >> support >> > >> > >>>>> Hive features, and make Hive to work with Flink. I agreed with >> Timo that on >> > >> > >>>>> starting with a smaller scope, so we can make progress faster. >> As for [6], >> > >> > >>>>> a proposal for DDL is already in progress, and will come after >> the unified >> > >> > >>>>> SQL connector API is done. For supporting Hive syntax, we might >> need to >> > >>>>> work with the Calcite community, and a recent effort called >> babel ( >> > >>>>> https://issues.apache.org/jira/browse/CALCITE-2280) in Calcite >> might >> > >>>>> help here. >> > >>>>> >> > >>>>> Thanks >> > >>>>> Shuyi >> > >>>>> >> > >>>>> On Wed, Oct 10, 2018 at 8:02 PM Zhang, Xuefu < >> > [hidden email]> >> > >>>>> wrote: >> > >>>>> Hi Fabian/Vno, >> > >>>>> >> > >> > >>>>> Thank you very much for your encouragement inquiry. Sorry that I >> didn't >> > >> > >>>>> see Fabian's email until I read Vino's response just now. >> (Somehow Fabian's >> > >>>>> went to the spam folder.) >> > >>>>> >> > >> > >>>>> My proposal contains long-term and short-terms goals. >> Nevertheless, the >> > >>>>> effort will focus on the following areas, including Fabian's >> list: >> > >>>>> >> > >>>>> 1. Hive metastore connectivity - This covers both read/write >> access, >> > >> > >>>>> which means Flink can make full use of Hive's metastore as its >> catalog (at >> > >>>>> least for the batch but can extend for streaming as well). >> > >> > >>>>> 2. Metadata compatibility - Objects (databases, tables, >> partitions, etc) >> > >> > >>>>> created by Hive can be understood by Flink and the reverse >> direction is >> > >>>>> true also. >> > >>>>> 3. Data compatibility - Similar to #2, data produced by Hive can >> be >> > >>>>> consumed by Flink and vise versa. >> > >> > >>>>> 4. Support Hive UDFs - For all Hive's native udfs, Flink either >> provides >> > >>>>> its own implementation or make Hive's implementation work in >> Flink. >> > >>>>> Further, for user created UDFs in Hive, Flink SQL should provide >> a >> > >> > >>>>> mechanism allowing user to import them into Flink without any >> code change >> > >>>>> required. >> > >>>>> 5. Data types - Flink SQL should support all data types that are >> > >>>>> available in Hive. >> > >>>>> 6. SQL Language - Flink SQL should support SQL standard (such as >> > >> > >>>>> SQL2003) with extension to support Hive's syntax and language >> features, >> > >>>>> around DDL, DML, and SELECT queries. >> > >> > >>>>> 7. SQL CLI - this is currently developing in Flink but more >> effort is >> > >>>>> needed. >> > >> > >>>>> 8. Server - provide a server that's compatible with Hive's >> HiverServer2 >> > >> > >>>>> in thrift APIs, such that HiveServer2 users can reuse their >> existing client >> > >>>>> (such as beeline) but connect to Flink's thrift server instead. >> > >> > >>>>> 9. JDBC/ODBC drivers - Flink may provide its own JDBC/ODBC >> drivers for >> > >>>>> other application to use to connect to its thrift server >> > >>>>> 10. Support other user's customizations in Hive, such as Hive >> Serdes, >> > >>>>> storage handlers, etc. >> > >> > >>>>> 11. Better task failure tolerance and task scheduling at Flink >> runtime. >> > >>>>> >> > >>>>> As you can see, achieving all those requires significant effort >> and >> > >> > >>>>> across all layers in Flink. However, a short-term goal could >> include only >> > >> > >>>>> core areas (such as 1, 2, 4, 5, 6, 7) or start at a smaller >> scope (such as >> > >>>>> #3, #6). >> > >>>>> >> > >> > >>>>> Please share your further thoughts. If we generally agree that >> this is >> > >> > >>>>> the right direction, I could come up with a formal proposal >> quickly and >> > >>>>> then we can follow up with broader discussions. >> > >>>>> >> > >>>>> Thanks, >> > >>>>> Xuefu >> > >>>>> >> > >>>>> >> > >>>>> >> > >>>>> >> ------------------------------------------------------------------ >> > >>>>> Sender:vino yang <[hidden email]> >> > >>>>> Sent at:2018 Oct 11 (Thu) 09:45 >> > >>>>> Recipient:Fabian Hueske <[hidden email]> >> > >>>>> Cc:dev <[hidden email]>; Xuefu <[hidden email] >> > >; user < >> > >>>>> [hidden email]> >> > >>>>> Subject:Re: [DISCUSS] Integrate Flink SQL well with Hive >> ecosystem >> > >>>>> >> > >>>>> Hi Xuefu, >> > >>>>> >> > >> > >>>>> Appreciate this proposal, and like Fabian, it would look better >> if you >> > >>>>> can give more details of the plan. >> > >>>>> >> > >>>>> Thanks, vino. >> > >>>>> >> > >>>>> Fabian Hueske <[hidden email]> 于2018年10月10日周三 下午5:27写道： >> > >>>>> Hi Xuefu, >> > >>>>> >> > >> > >>>>> Welcome to the Flink community and thanks for starting this >> discussion! >> > >>>>> Better Hive integration would be really great! >> > >>>>> Can you go into details of what you are proposing? I can think >> of a >> > >>>>> couple ways to improve Flink in that regard: >> > >>>>> >> > >>>>> * Support for Hive UDFs >> > >>>>> * Support for Hive metadata catalog >> > >>>>> * Support for HiveQL syntax >> > >>>>> * ??? >> > >>>>> >> > >>>>> Best, Fabian >> > >>>>> >> > >>>>> Am Di., 9. Okt. 2018 um 19:22 Uhr schrieb Zhang, Xuefu < >> > >>>>> [hidden email]>: >> > >>>>> Hi all, >> > >>>>> >> > >>>>> Along with the community's effort, inside Alibaba we have >> explored >> > >> > >>>>> Flink's potential as an execution engine not just for stream >> processing but >> > >>>>> also for batch processing. We are encouraged by our findings and >> have >> > >> > >>>>> initiated our effort to make Flink's SQL capabilities >> full-fledged. When >> > >> > >>>>> comparing what's available in Flink to the offerings from >> competitive data >> > >> > >>>>> processing engines, we identified a major gap in Flink: a well >> integration >> > >> > >>>>> with Hive ecosystem. This is crucial to the success of Flink SQL >> and batch >> > >> > >>>>> due to the well-established data ecosystem around Hive. >> Therefore, we have >> > >> > >>>>> done some initial work along this direction but there are still >> a lot of >> > >>>>> effort needed. >> > >>>>> >> > >>>>> We have two strategies in mind. The first one is to make Flink >> SQL >> > >> > >>>>> full-fledged and well-integrated with Hive ecosystem. This is a >> similar >> > >> > >>>>> approach to what Spark SQL adopted. The second strategy is to >> make Hive >> > >> > >>>>> itself work with Flink, similar to the proposal in [1]. Each >> approach bears >> > >> > >>>>> its pros and cons, but they don’t need to be mutually exclusive >> with each >> > >>>>> targeting at different users and use cases. We believe that both >> will >> > >>>>> promote a much greater adoption of Flink beyond stream >> processing. >> > >>>>> >> > >>>>> We have been focused on the first approach and would like to >> showcase >> > >> > >>>>> Flink's batch and SQL capabilities with Flink SQL. However, we >> have also >> > >>>>> planned to start strategy #2 as the follow-up effort. >> > >>>>> >> > >> > >>>>> I'm completely new to Flink(, with a short bio [2] below), >> though many >> > >> > >>>>> of my colleagues here at Alibaba are long-time contributors. >> Nevertheless, >> > >> > >>>>> I'd like to share our thoughts and invite your early feedback. >> At the same >> > >> > >>>>> time, I am working on a detailed proposal on Flink SQL's >> integration with >> > >>>>> Hive ecosystem, which will be also shared when ready. >> > >>>>> >> > >>>>> While the ideas are simple, each approach will demand significant >> > >> > >>>>> effort, more than what we can afford. Thus, the input and >> contributions >> > >>>>> from the communities are greatly welcome and appreciated. >> > >>>>> >> > >>>>> Regards, >> > >>>>> >> > >>>>> >> > >>>>> Xuefu >> > >>>>> >> > >>>>> References: >> > >>>>> >> > >>>>> [1] https://issues.apache.org/jira/browse/HIVE-10712 >> > >> > >>>>> [2] Xuefu Zhang is a long-time open source veteran, worked or >> working on >> > >>>>> many projects under Apache Foundation, of which he is also an >> honored >> > >> > >>>>> member. About 10 years ago he worked in the Hadoop team at Yahoo >> where the >> > >> > >>>>> projects just got started. Later he worked at Cloudera, >> initiating and >> > >> > >>>>> leading the development of Hive on Spark project in the >> communities and >> > >> > >>>>> across many organizations. Prior to joining Alibaba, he worked >> at Uber >> > >> > >>>>> where he promoted Hive on Spark to all Uber's SQL on Hadoop >> workload and >> > >>>>> significantly improved Uber's cluster efficiency. >> > >>>>> >> > >>>>> >> > >>>>> >> > >>>>> >> > >>>>> -- >> > >> > >>>>> "So you have to trust that the dots will somehow connect in your >> future." >> > >>>>> >> > >>>>> >> > >>>>> -- >> > >> > >>>>> "So you have to trust that the dots will somehow connect in your >> future." >> > >>>>> >> > >> > >> >> |
«
Return to (DEPRECATED) Apache Flink Mailing List archive.
|
1 view|%1 views
| Free forum by Nabble | Edit this page |