[
https://issues.apache.org/jira/browse/CASSANDRA-6861?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13990601#comment-13990601
]
Benedict commented on CASSANDRA-6861:
-------------------------------------
Performance graphs look nice - much tighter variance.
Couple of nits with the patch:
1) It would be nice, since we effectively create a static allocator, to
reference it from a static context instead of grabbing it from the buffer we've
been passed by Netty
2) I'm a little concerned we aren't tidying up in the case of exceptions - this
is probably not a major issue since netty uses finalizers and exceptions should
be rare, so there'll be little resource leakage, but it would be good to at
least tidy up the frame in the case that we have to log an exception. This
won't catch all possible paths (e.g. exceptions before we finish setting the
frame) but it's probably close enough.
> Optimise our Netty 4 integration
> --------------------------------
>
> Key: CASSANDRA-6861
> URL: https://issues.apache.org/jira/browse/CASSANDRA-6861
> Project: Cassandra
> Issue Type: Improvement
> Components: Core
> Reporter: Benedict
> Assignee: T Jake Luciani
> Priority: Minor
> Labels: performance
> Fix For: 2.1 rc1
>
>
> Now we've upgraded to Netty 4, we're generating a lot of garbage that could
> be avoided, so we should probably stop that. Should be reasonably easy to
> hook into Netty's pooled buffers, returning them to the pool once a given
> message is completed.
--
This message was sent by Atlassian JIRA
(v6.2#6252)
