[ 
https://issues.apache.org/jira/browse/CASSANDRA-20975?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=18084589#comment-18084589
 ] 

Joey Lynch edited comment on CASSANDRA-20975 at 5/30/26 5:35 PM:
-----------------------------------------------------------------

Left some feedback on 
[https://github.com/apache/cassandra/pull/4804#pullrequestreview-4395475293]

Overall I like the approach, I think we just need to tighten and test the 
failure cases a little more. I am slightly concerned with what happens if a 
compressor throws during the provide call, or if it doesn't provide identical 
output to the default. I think the former can be tested, and the latter maybe 
the smoke test idea? 


was (Author: jolynch):
Left some feedback on 
[https://github.com/apache/cassandra/pull/4804#pullrequestreview-4395475293]

Overall I like the approach, I think we just need to tighten and test the 
failure cases a little more. I am slightly concerned with what happens if a 
compressor throws during the provide call, or if it doesn't provide identical 
output to the default. 

> CEP-49: Hardware-accelerated compression
> ----------------------------------------
>
>                 Key: CASSANDRA-20975
>                 URL: https://issues.apache.org/jira/browse/CASSANDRA-20975
>             Project: Apache Cassandra
>          Issue Type: New Feature
>          Components: Feature/Compression
>            Reporter: shylaja kokoori
>            Assignee: shylaja kokoori
>            Priority: Normal
>             Fix For: 6.x
>
>          Time Spent: 5h 20m
>  Remaining Estimate: 0h
>
> CEP-49 proposes improving performance of compress/decompress operations with 
> hardware-based accelerators. 
> Based on feedback from the mailing list, creating this ticket to enable 
> loading accelerated compression as a pluggable service provider, if jar file 
> is available in the classpath.



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

---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]

Reply via email to