beam-commits mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Aljoscha Krettek (JIRA)" <>
Subject [jira] [Commented] (BEAM-1612) Support real Bundle in Flink runner
Date Fri, 03 Mar 2017 13:26:45 GMT


Aljoscha Krettek commented on BEAM-1612:

Yes, I think finding a solution for bundling is very important.

Right now, I'm afraid we have to manually induce bundling using either count or time (or both).
Making bundling dependent on snapshotting doesn't work because Flink operators cannot (right
now) emit data while snapshotting while a Beam {{DoFn}} can emit data in the {{@FinishBundle}}

> Support real Bundle in Flink runner
> -----------------------------------
>                 Key: BEAM-1612
>                 URL:
>             Project: Beam
>          Issue Type: Improvement
>          Components: runner-flink
>            Reporter: Jingsong Lee
>            Assignee: Jingsong Lee
> The Bundle is very important in the beam model. Users can use the bundle to flush buffer,
can reuse many heavyweight resources in a bundle. Most IO plugins use the bundle to flush.

> Moreover, FlinkRunner can also use Bundle to reduce access to the FlinkState, such as
first placed in JavaHeap, flush into RocksDbState when invoke finishBundle , this can reduce
the number of serialization.
> But now FlinkRunner calls the finishBundle every processElement. We need support real
> I think we can have the following implementations:
> 1.Invoke finishBundle and next startBundle in {{snapshot}} of Flink. But sometimes this
"Bundle" maybe too big. This depends on the user's checkpoint configuration.
> 2.Manually control the size of the bundle. The half-bundle will be flushed to a full-bundle
by count or eventTime or processTime or {{snapshot}}. We do not need to wait, just call the
startBundle and finishBundle at the right time.

This message was sent by Atlassian JIRA

View raw message