You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@spark.apache.org by "jiangjiguang0719 (Jira)" <ji...@apache.org> on 2023/03/07 08:56:00 UTC

[jira] [Created] (SPARK-42696) Speed up parquet reading with Java Vector API

jiangjiguang0719 created SPARK-42696:
----------------------------------------

             Summary: Speed up parquet reading with Java Vector API
                 Key: SPARK-42696
                 URL: https://issues.apache.org/jira/browse/SPARK-42696
             Project: Spark
          Issue Type: New Feature
          Components: Input/Output
    Affects Versions: 3.5.0
            Reporter: jiangjiguang0719


Parquet has supported use Java 17 Vector API to perform bit-unpacking to enjoy 4x ~ 8x performance gain in microbenchmark.

I have finished the TPC-H(SF100) benchmark with spark integrated parquet optimization, each SQL has a different performance gain,  Q6 can reach up 11%

 

Please assign it to me, I will summit a PR, thanks!



--
This message was sent by Atlassian Jira
(v8.20.10#820010)

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