You are viewing a plain text version of this content. The canonical link for it is here.

Posted to issues@spark.apache.org by "Yesheng Ma (JIRA)" <ji...@apache.org> on 2019/06/19 22:19:00 UTC

[jira] [Created] (SPARK-28113) Lazy val performance pitfall on Spark LogicalPlan's output method

Yesheng Ma created SPARK-28113:
----------------------------------

             Summary: Lazy val performance pitfall on Spark LogicalPlan's output method
                 Key: SPARK-28113
                 URL: https://issues.apache.org/jira/browse/SPARK-28113
             Project: Spark
          Issue Type: Improvement
          Components: SQL
    Affects Versions: 3.0.0
            Reporter: Yesheng Ma


The original {{output}} implementation in a few QueryPlan sub-classes are methods, which means unnecessary re-computation can happen at times. This PR resolves this problem by making these method lazy vals.

We benchmarked this optimization on TPC-DS. In the benchmark, we warmed up the queries 5 iterations and then took the average of 5 runs. Results showed that this micro-optimization can improve the end-to-end planning time by 9.3%.



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)

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