You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@cassandra.apache.org by "Jonathan Ellis (JIRA)" <ji...@apache.org> on 2011/02/17 03:09:24 UTC
[jira] Commented: (CASSANDRA-2178) Memtable Flush writers doesn't
actually flush in parallel
[ https://issues.apache.org/jira/browse/CASSANDRA-2178?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12995633#comment-12995633 ]
Jonathan Ellis commented on CASSANDRA-2178:
-------------------------------------------
Wouldn't setting {core=1, max=flush_writers, queue=SynchronousQueue) be better since it allows the pool size to adjust as needed?
> Memtable Flush writers doesn't actually flush in parallel
> ---------------------------------------------------------
>
> Key: CASSANDRA-2178
> URL: https://issues.apache.org/jira/browse/CASSANDRA-2178
> Project: Cassandra
> Issue Type: Bug
> Affects Versions: 0.7.2, 0.8
> Reporter: Chris Goffinet
> Assignee: Chris Goffinet
> Priority: Minor
> Fix For: 0.8
>
> Attachments: 0001-Set-the-core-min-pool-size-to-memtable_flush_writers.patch
>
>
> The flushWriter JMXEnabledThreadPoolExecutor sets the core pool min to 1, and sets the LBQ to DatabaseDescriptor.getFlushWriters(). Increasing memtable_flush_writers should allow us to flush more in parallel. The pool will not grow until LBQ fills up to DatabaseDescriptor.getFlushWriters().
--
This message is automatically generated by JIRA.
-
For more information on JIRA, see: http://www.atlassian.com/software/jira