[
https://issues.apache.org/jira/browse/FLINK-31144?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17692527#comment-17692527
]
Zhu Zhu commented on FLINK-31144:
---------------------------------
I suspect the slowness is caused by the N^2 complexity to compute the input
locations when there are N upstream task and N downstream tasks.
If so, as long as N is not too large, e.g. not larger than
MAX_DISTINCT_CONSUMERS_TO_CONSIDER=8, the cost of input location computation
should be acceptable.
Also, if there are too many distinct consumers, input locality would make none
sense.
As [~huwh] mentioned, it's hard for users to decides a proper threshold for
each job. It's also inconvenient if users had to decide whether to enable input
locality or not. Therefore, I prefer to let Flink decide it automatically for
users, like the proposed change above.
> Slow scheduling on large-scale batch jobs
> ------------------------------------------
>
> Key: FLINK-31144
> URL: https://issues.apache.org/jira/browse/FLINK-31144
> Project: Flink
> Issue Type: Bug
> Components: Runtime / Coordination
> Reporter: Julien Tournay
> Priority: Major
> Attachments: flink-1.17-snapshot-1676473798013.nps,
> image-2023-02-21-10-29-49-388.png
>
>
> When executing a complex job graph at high parallelism
> `DefaultPreferredLocationsRetriever.getPreferredLocationsBasedOnInputs` can
> get slow and cause long pauses where the JobManager becomes unresponsive and
> all the taskmanagers just wait. I've attached a VisualVM snapshot to
> illustrate the problem.[^flink-1.17-snapshot-1676473798013.nps]
> At Spotify we have complex jobs where this issue can cause batch "pause" of
> 40+ minutes and make the overall execution 30% slower or more.
> More importantly this prevent us from running said jobs on larger cluster as
> adding resources to the cluster worsen the issue.
> We have successfully tested a modified Flink version where
> `DefaultPreferredLocationsRetriever.getPreferredLocationsBasedOnInputs` was
> completely commented and simply returns an empty collection and confirmed it
> solves the issue.
> In the same spirit as a recent change
> ([https://github.com/apache/flink/blob/43f419d0eccba86ecc8040fa6f521148f1e358ff/flink-runtime/src/main/java/org/apache/flink/runtime/scheduler/DefaultPreferredLocationsRetriever.java#L98-L102)]
> there could be a mechanism in place to detect when Flink run into this
> specific issue and just skip the call to `getInputLocationFutures`
> [https://github.com/apache/flink/blob/43f419d0eccba86ecc8040fa6f521148f1e358ff/flink-runtime/src/main/java/org/apache/flink/runtime/scheduler/DefaultPreferredLocationsRetriever.java#L105-L108.]
> I'm not familiar enough with the internals of Flink to propose a more
> advanced fix, however it seems like a configurable threshold on the number of
> consumer vertices above which the preferred location is not computed would
> do. If this solution is good enough, I'd be happy to submit a PR.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
