[
https://issues.apache.org/jira/browse/CASSANDRA-16465?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17331866#comment-17331866
]
Nicolai Vest commented on CASSANDRA-16465:
------------------------------------------
[~jaid] We use only TWCS and LCS with frequent read and write operations in our
setup and regression can only be seen for the TWCS. However, STCS might be
effected too but you should do your own testing to find out.
What we observe are increased SSTable reads, especially for 95th percentile and
higher, and increased read latency up to twice as high as before.
We observe the same regression for 3.11.7, 3.11.8, 3.11.9. We did not test
3.11.10 yet but could do that easily, if you ask for.
> Increased Read Latency With Cassandra >= 3.11.7
> -----------------------------------------------
>
> Key: CASSANDRA-16465
> URL: https://issues.apache.org/jira/browse/CASSANDRA-16465
> Project: Cassandra
> Issue Type: Bug
> Components: Legacy/Local Write-Read Paths
> Reporter: Ahmed ELJAMI
> Priority: Normal
> Fix For: 3.11.x, 4.0.x
>
> Attachments: Screenshot from 2021-04-24 20-50-47.png
>
>
> After upgrading Cassandra from 3.11.3 to 3.11.9, Cassandra read latency 99%
> increased significantly. Getting back to 3.11.3 immediately fixed the issue.
> I have observed "SStable reads" increases after upgrading to 3.11.9.
> The same behavior was observed by some other users:
> [https://www.mail-archive.com/[email protected]/msg61247.html]
> According to Paulo Motta's comment, this behavior may be caused by
> https://issues.apache.org/jira/browse/CASSANDRA-15690 which was introduced on
> 3.11.7 and removed an optimization that may cause a correctness issue when
> there are partition deletions.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
