[GitHub] [flink-benchmarks] wsry commented on pull request #44: [FLINK-25704] Fix the blocking partition benchmark regression caused by FLINK-25637

[GitBox <gi...@apache.org>]

wsry commented on pull request #44:
URL: https://github.com/apache/flink-benchmarks/pull/44#issuecomment-1022004795


   > NETWORK_SORT_SHUFFLE_MIN_PARALLELISM
   
   @pnowojski Thanks for the review.
   
   ```NETWORK_SORT_SHUFFLE_MIN_PARALLELISM``` is a config option to config which blocking shuffle implementation to use, either the hash-based one or the sort-based one. If task parallelism is smaller than this config value, the hash-based blocking shuffle will be used, otherwise, the sort-based blocking shuffle will be used. Previously, the default value of this config option is Integer.MAX_VALUE, which mean hash-shuffle will be used by default, FLINK-25637 changed the default value to 1, which means sort-shuffle will be used by default. The changes in this PR restore the previous default value for the corresponding benchmark cases.


-- 
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.

To unsubscribe, e-mail: issues-unsubscribe@flink.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org