You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@spark.apache.org by "Michael Zhang (Jira)" <ji...@apache.org> on 2021/07/20 22:31:00 UTC

[jira] [Created] (SPARK-36234) Consider mapper location and shuffle block size in OptimizeLocalShuffleReader

Michael Zhang created SPARK-36234:
-------------------------------------

             Summary: Consider mapper location and shuffle block size in OptimizeLocalShuffleReader
                 Key: SPARK-36234
                 URL: https://issues.apache.org/jira/browse/SPARK-36234
             Project: Spark
          Issue Type: Improvement
          Components: SQL
    Affects Versions: 3.1.2
            Reporter: Michael Zhang


This is a follow-up to SPARK-36105 (OptimizeLocalShuffleReader support reading data of multiple mappers in one task). We should consider using the mapper locations along with shuffle block size when coalescing mappers (specifically in events where there are more mappers than there is parallelism.



--
This message was sent by Atlassian Jira
(v8.3.4#803005)

---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscribe@spark.apache.org
For additional commands, e-mail: issues-help@spark.apache.org