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

Posted to commits@hudi.apache.org by "Ethan Guo (Jira)" <ji...@apache.org> on 2023/02/16 07:13:00 UTC

[jira] [Updated] (HUDI-5347) Fix Merge Into performance traps

     [ https://issues.apache.org/jira/browse/HUDI-5347?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Ethan Guo updated HUDI-5347:
----------------------------
    Fix Version/s: 0.13.0

> Fix Merge Into performance traps
> --------------------------------
>
>                 Key: HUDI-5347
>                 URL: https://issues.apache.org/jira/browse/HUDI-5347
>             Project: Apache Hudi
>          Issue Type: Bug
>          Components: spark-sql
>    Affects Versions: 0.12.1
>            Reporter: Alexey Kudinkin
>            Assignee: Alexey Kudinkin
>            Priority: Blocker
>              Labels: performance, pull-request-available
>             Fix For: 0.12.2, 0.13.0
>
>
> Merge Into currently relies on SparkSqlTypedRecord abstraction doing for *every* field of *every* record lookup in the cache keyed by Avro's Schema incurring Schema.equals (bearing non-trivial overhead), leading to 95% of the compute being wasted on it.



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