[ 
https://issues.apache.org/jira/browse/CASSANDRA-16465?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17334381#comment-17334381
 ] 

Nicolai Vest commented on CASSANDRA-16465:
------------------------------------------

This is showing the impact of the performance regression for the different 
versions for a selected TWCS table in our test setup (three C* nodes in one 
cluster and with just little load). I marked the dates I installed the 
different versions. For versions 3.11.7 to 3.11.10 SSTable reads are related to 
the number of total SSTables for that table. For us the impact becomes even 
more dramatic when the load increases.

!image-2021-04-28-02-11-27-298.png|width=486,height=460!

 

[~stefan.miklosovic] 

Regarding -CASSANDRA-15690-  versions affected are 
[3.0.21|https://issues.apache.org/jira/issues/?jql=project+%3D+CASSANDRA+AND+fixVersion+%3D+3.0.21],
 
[3.11.7|https://issues.apache.org/jira/issues/?jql=project+%3D+CASSANDRA+AND+fixVersion+%3D+3.11.7],
 
[4.0|https://issues.apache.org/jira/issues/?jql=project+%3D+CASSANDRA+AND+fixVersion+%3D+4.0],
 
[4.0-beta1|https://issues.apache.org/jira/issues/?jql=project+%3D+CASSANDRA+AND+fixVersion+%3D+4.0-beta1],
 ... and newer I guess. 

> 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, 
> image-2021-04-28-02-11-27-298.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]

Reply via email to