flink-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Stefan Richter <s.rich...@data-artisans.com>
Subject Re: Managed operator state treating state of all parallel operators as the same
Date Mon, 03 Jul 2017 16:21:17 GMT
Hi,

I think you can do this already, please take a look at OperatorStateStore::getUnionListState(…).
You could add state to this only one one parallel operator instance, e.g. by checking the
subtask index.

Best,
Stefan

> Am 03.07.2017 um 18:14 schrieb Fabian Hueske <fhueske@gmail.com>:
> 
> Hi Gerard,
> 
> such a type of state is not yet supported, but on the roadmap. 
> The feature is called Broadcast State. A description is in JIRA [1].
> 
> Best, Fabian
> 
> [1] https://issues.apache.org/jira/browse/FLINK-4940 <https://issues.apache.org/jira/browse/FLINK-4940>
> 
> 2017-07-03 17:50 GMT+02:00 gerardg <gerard@talaia.io <mailto:gerard@talaia.io>>:
> Hello,
> 
> Is it possible to have managed operator state where all the parallel
> operators know that they have the same state? Therefore, it would be only
> snapshotted by one of them and in case of failure (or rescaling) all would
> use that snapshot.
> 
> For the rescaling case I guess I could use union redistribution and just
> keep one of the elements of the list, but given the size of the state it
> would be more efficient if this can be avoided.
> 
> Thanks,
> 
> Gerard
> 
> 
> 
> --
> View this message in context: http://apache-flink-user-mailing-list-archive.2336050.n4.nabble.com/Managed-operator-state-treating-state-of-all-parallel-operators-as-the-same-tp14102.html
<http://apache-flink-user-mailing-list-archive.2336050.n4.nabble.com/Managed-operator-state-treating-state-of-all-parallel-operators-as-the-same-tp14102.html>
> Sent from the Apache Flink User Mailing List archive. mailing list archive at Nabble.com.
> 


Mime
View raw message