You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@hudi.apache.org by "Sagar Sumit (Jira)" <ji...@apache.org> on 2022/04/06 16:43:00 UTC

[jira] [Commented] (HUDI-3791) Test perf for point looks up for bloom filter and col stats partition

    [ https://issues.apache.org/jira/browse/HUDI-3791?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17518298#comment-17518298 ] 

Sagar Sumit commented on HUDI-3791:
-----------------------------------

Item#4 : Point lookup on column_stats works

I verified that not only it works but also that only files in `./hoodie/metadata/` are begin fetched

> Test perf for point looks up for bloom filter and col stats partition
> ---------------------------------------------------------------------
>
>                 Key: HUDI-3791
>                 URL: https://issues.apache.org/jira/browse/HUDI-3791
>             Project: Apache Hudi
>          Issue Type: Task
>          Components: metadata
>            Reporter: sivabalan narayanan
>            Assignee: Sagar Sumit
>            Priority: Blocker
>             Fix For: 0.11.0
>
>
> # Enable col stats and bloom filter for 100k+ files tables and ensure upserts and query works. (w/o point look ups)
>  # Enable col stats and bloom filter for 100k+ files tables and ensure upserts and query works. (w point look ups)
>  # w/ and w/o point look ups, get a sense perf difference. Try to chase down any fixes that we can spot atleast to get point look ups on par w/ full scan. 
>  # Micro benchmark for sanity check that point look ups on col stats works. 
>  
>  



--
This message was sent by Atlassian Jira
(v8.20.1#820001)