Unique ID of an operator

[Deepak Jha]

Hi,
I have a use case where I need to get UniqueId of an operator inside a
stream. DataStream's getId() returns the ID of the stream but I have
operator partitioned (say partitionByHash) inside the datastream. So I
would like to get unique ID for each operator working in parallel. Is there
a way in which I can get it ?
Also, does DataStream's getId generates different ID in case i have
parallelism > 1 ?


--
Thanks,
Deepak

[Robert Metzger]

Hi,

the runtime context has the following method:

int subtask = getRuntimeContext().getIndexOfThisSubtask();

The RuntimeContext is available in each Rich* function.


On Fri, Apr 8, 2016 at 11:46 PM, Deepak Jha <[hidden email]> wrote:

> Hi,
> I have a use case where I need to get UniqueId of an operator inside a
> stream. DataStream's getId() returns the ID of the stream but I have
> operator partitioned (say partitionByHash) inside the datastream. So I
> would like to get unique ID for each operator working in parallel. Is there
> a way in which I can get it ?
> Also, does DataStream's getId generates different ID in case i have
> parallelism > 1 ?
>
>
> --
> Thanks,
> Deepak
>

[Deepak Jha]

Hi Robert,
Thanks.

On Sat, Apr 9, 2016 at 12:55 AM, Robert Metzger <[hidden email]
<javascript:_e(%7B%7D,'cvml','[hidden email]');>> wrote:

> Hi,
>
> the runtime context has the following method:
>
> int subtask = getRuntimeContext().getIndexOfThisSubtask();
>
> The RuntimeContext is available in each Rich* function.
>
>
> On Fri, Apr 8, 2016 at 11:46 PM, Deepak Jha <[hidden email]
> <javascript:_e(%7B%7D,'cvml','[hidden email]');>> wrote:
>
> > Hi,
> > I have a use case where I need to get UniqueId of an operator inside a
> > stream. DataStream's getId() returns the ID of the stream but I have
> > operator partitioned (say partitionByHash) inside the datastream. So I
> > would like to get unique ID for each operator working in parallel. Is
> there
> > a way in which I can get it ?
> > Also, does DataStream's getId generates different ID in case i have
> > parallelism > 1 ?
> >
> >
> > --
> > Thanks,
> > Deepak
> >
>

--
Thanks,
Deepak Jha



--
Sent from Gmail Mobile

[jhon22]

For readers exploring practical technology discussions and online resources, the archived Apache Flink conversation offers useful context around developer-focused problem solving. For a change of pace from technical topics, No Boundaries provides an everyday fashion collection focused on versatile pieces that are easy to wear and built for regular rotation, from casual basics to dependable statement styles