[ 
https://issues.apache.org/jira/browse/CASSANDRA-15393?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17185732#comment-17185732
 ] 

Michael Semb Wever commented on CASSANDRA-15393:
------------------------------------------------

We have been strict on the 4.0 feature freeze and are now in the 4.0-beta phase.

With the agreed upon release lifecycle: I struggle with the justification for 
putting this into 4.0.0, regardless of the improvements the patch brings.

If it met the criteria to go into 3.11.x then 4.0-beta is obvious. But it's not 
a bug, and neither a stability/performance improvement on a 4.0 introduced 
feature. Isn't this how we should be triaging all tickets from here to 4.0-rc.
{quote}The C* project has a problem with favoring rewrites in favor of 
incremental improvements.
{quote}
Totally agree with you [~bdeggleston]. I would definitely vote for the 
incremental approach, but starting in 4.0.x or 4.1. Especially after we have a 
better QA/test process spec'd out during the 4.0-beta phase.
{quote}The 4.0.x line is going to be around for a while…
{quote}
[~maedhroz], that's not an argument in itself to waiver this patch into 
4.0-beta, and sets the precedence for feature creep.

Can I suggest this needs to be raised on dev ML to get a waiver for 4.0-beta?

> Add byte array backed cells
> ---------------------------
>
>                 Key: CASSANDRA-15393
>                 URL: https://issues.apache.org/jira/browse/CASSANDRA-15393
>             Project: Cassandra
>          Issue Type: Sub-task
>          Components: Local/Compaction
>            Reporter: Blake Eggleston
>            Assignee: Blake Eggleston
>            Priority: Normal
>             Fix For: 4.0-beta
>
>          Time Spent: 20m
>  Remaining Estimate: 0h
>
> We currently materialize all values as on heap byte buffers. Byte buffers 
> have a fairly high overhead given how frequently they’re used, and on the 
> compaction and local read path we don’t do anything that needs them. Use of 
> byte buffer methods only happens on the coordinator. Using cells that are 
> backed by byte arrays instead in these situations reduces compaction and read 
> garbage up to 22% in many cases.



--
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