[
https://issues.apache.org/jira/browse/CASSANDRA-18734?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17762448#comment-17762448
]
Mike Adamson commented on CASSANDRA-18734:
------------------------------------------
Final test runs are here:
|[JDK11|https://app.circleci.com/pipelines/github/mike-tr-adamson/cassandra/278/workflows/2b0d0a26-c175-4ef2-bf7d-6df982f9dfa6]|
|[JDK17|https://app.circleci.com/pipelines/github/mike-tr-adamson/cassandra/278/workflows/1643c3f9-b55f-4285-bfce-0c792fc36a71]|
There is a failure on the 17 test run that looks unrelated.
> SAI result retriever is filtering too many rows
> -----------------------------------------------
>
> Key: CASSANDRA-18734
> URL: https://issues.apache.org/jira/browse/CASSANDRA-18734
> Project: Cassandra
> Issue Type: Bug
> Components: Feature/SAI
> Reporter: Mike Adamson
> Assignee: Mike Adamson
> Priority: Normal
> Attachments: flamegraph.svg, image-2023-08-08-13-48-19-701.png,
> image-2023-08-08-13-49-19-576.png
>
>
> Performance tests on SAI have shown that the number of rows being filtered
> for wide row partitions is more than expected. For a 10k row/partition test -
> limit 10, the following has been observed.
> !image-2023-08-08-13-49-19-576.png|width=952,height=368!
> This is not the expected outcome of row-awareness and needs investigating.
> The number of rows read should, roughly, match the number of partitions read.
>
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: commits-unsubscr...@cassandra.apache.org
For additional commands, e-mail: commits-h...@cassandra.apache.org
