impala-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Thomas Tauber-Marshall (JIRA)" <>
Subject [jira] [Created] (IMPALA-5649) Use partial sort for HDFS tables
Date Tue, 11 Jul 2017 23:54:01 GMT
Thomas Tauber-Marshall created IMPALA-5649:

             Summary: Use partial sort for HDFS tables
                 Key: IMPALA-5649
             Project: IMPALA
          Issue Type: Improvement
          Components: Frontend
            Reporter: Thomas Tauber-Marshall

A change currently in review (IMPALA-5498) is adding the ability to do partial sorts, where
the input is divided up into batches each of which is sorted individually, allowing us to
avoid spilling when sorting large inputs.

The initial use case for it is inserts into Kudu tables, but it could also be used to improve
the performance of sorted inserts into hdfs tables.

This message was sent by Atlassian JIRA

View raw message