You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@cassandra.apache.org by "Benedict Elliott Smith (Jira)" <ji...@apache.org> on 2020/01/22 10:43:00 UTC
[jira] [Commented] (CASSANDRA-15516) Failed partition updates
should not consume memtable space
[ https://issues.apache.org/jira/browse/CASSANDRA-15516?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17020950#comment-17020950 ]
Benedict Elliott Smith commented on CASSANDRA-15516:
----------------------------------------------------
For those interested, a performance comparison of this work against 3.0 can be found on CASSANDRA-15511, which builds on this patch to be able to remove the partition lock.
> Failed partition updates should not consume memtable space
> ----------------------------------------------------------
>
> Key: CASSANDRA-15516
> URL: https://issues.apache.org/jira/browse/CASSANDRA-15516
> Project: Cassandra
> Issue Type: Improvement
> Components: Local/Memtable
> Reporter: Benedict Elliott Smith
> Priority: Normal
>
> Moving my proposed patch for CASSANDRA-15367 here. Contended memtable updates currently waste memtable memory each time they fail to apply. This patch avoids this problem by ensuring we only copy data once onto the memtable, and reconcile this new data for all future attempts to insert.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
---------------------------------------------------------------------
To unsubscribe, e-mail: commits-unsubscribe@cassandra.apache.org
For additional commands, e-mail: commits-help@cassandra.apache.org