You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@flink.apache.org by GitBox <gi...@apache.org> on 2019/12/16 16:27:36 UTC

[GitHub] [flink] pnowojski commented on issue #10529: [FLINK-15171] [serialization] fix performance regression caused by too many buffer allocations on string serialization

pnowojski commented on issue #10529: [FLINK-15171] [serialization] fix performance regression caused by too many buffer allocations on string serialization
URL: https://github.com/apache/flink/pull/10529#issuecomment-566135495
 
 
   @shuttie , yes, I think the GC pressure is the prime suspect here. For example the brief blimp in the performance on the timeline around http://codespeed.dak8s.net:8000/timeline/?ben=serializerTuple&env=2 was caused by some memory related configuration changes (part of another investigation). I've revert those changes after a couple of runs, to not skew benchmark results for this regression.
   
   Once you have provided an update, please ping me or @rkhachatryan so we can re-trigger the benchmark request run :)

----------------------------------------------------------------
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:
users@infra.apache.org


With regards,
Apache Git Services