[
https://issues.apache.org/jira/browse/BEAM-8151?focusedWorklogId=343826&page=com.atlassian.jira.plugin.system.issuetabpanels:worklog-tabpanel#worklog-343826
]
ASF GitHub Bot logged work on BEAM-8151:
----------------------------------------
Author: ASF GitHub Bot
Created on: 14/Nov/19 23:02
Start Date: 14/Nov/19 23:02
Worklog Time Spent: 10m
Work Description: lukecwik commented on issue #10091: [BEAM-8151] Create
1000 SdkWorker objects instead of the current limit of 12.
URL: https://github.com/apache/beam/pull/10091#issuecomment-554124035
I tried generating the workers on demand as can be seen in
https://github.com/lukecwik/incubator-beam/commit/c743564d3dd029260c0d8a5e703856c9ddf2650a
but I ran into lots of test issues which assumed that there was only a single
SdkWorker and instead of disabling those tests I choose to go with this short
term solution to fix a concurrency limit issue with the Python SDK.
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org
Issue Time Tracking
-------------------
Worklog Id: (was: 343826)
Time Spent: 12h 20m (was: 12h 10m)
> Allow the Python SDK to use many many threads
> ---------------------------------------------
>
> Key: BEAM-8151
> URL: https://issues.apache.org/jira/browse/BEAM-8151
> Project: Beam
> Issue Type: Improvement
> Components: sdk-py-core, sdk-py-harness
> Reporter: Luke Cwik
> Assignee: Luke Cwik
> Priority: Major
> Time Spent: 12h 20m
> Remaining Estimate: 0h
>
> We need to use a thread pool which shrinks the number of active threads when
> they are not being used.
>
> This is to prevent any stuckness issues related to a runner scheduling more
> work items then there are "work" threads inside the SDK harness.
>
> By default the control plane should have all "requests" being processed in
> parallel and the runner is responsible for not overloading the SDK with too
> much work.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
