[
https://issues.apache.org/jira/browse/BEAM-11403?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17298290#comment-17298290
]
Boyuan Zhang commented on BEAM-11403:
-------------------------------------
Sorry for the late.
One last improvement I can come up with is to make checkpoint frequency can be
configured by pipeline author:
https://docs.google.com/document/d/18jNLtTyyApx0N2ytp1ytOMmUPLouj2h08N3-4SyWGgQ/edit?usp=sharing.
>From what Kaymak reported:
>https://lists.apache.org/thread.html/r3972e2b6016f59af4bea7b3729f5467d0201007b6f217512914c2fb6%40%3Cuser.beam.apache.org%3E,
> it seems like watermark is still held back with
>https://github.com/apache/beam/pull/14013.
> Unbounded SDF wrapper causes performance regression on DirectRunner
> -------------------------------------------------------------------
>
> Key: BEAM-11403
> URL: https://issues.apache.org/jira/browse/BEAM-11403
> Project: Beam
> Issue Type: Bug
> Components: runner-direct, sdk-java-core
> Reporter: Boyuan Zhang
> Priority: P2
> Time Spent: 4.5h
> Remaining Estimate: 0h
>
> There is a significant performance regression when switching from
> UnboundedSource to Unbounded SDF wrapper. So far there are 2 IOs reported:
> * Pubsub Read:
> https://lists.apache.org/thread.html/re6b0941a8b4951293a0327ce9b25e607cafd6e45b69783f65290edee%40%3Cdev.beam.apache.org%3E
> * Kafka Read: https://the-asf.slack.com/archives/C9H0YNP3P/p1606155042346600
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
