[jira] [Commented] (CASSANDRA-13290) Optimizing very small repair streams

["Jeff Jirsa (JIRA)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/CASSANDRA-13290?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15892947#comment-15892947 ] 

Jeff Jirsa commented on CASSANDRA-13290:
----------------------------------------

See also CASSANDRA-8911



> Optimizing very small repair streams
> ------------------------------------
>
>                 Key: CASSANDRA-13290
>                 URL: https://issues.apache.org/jira/browse/CASSANDRA-13290
>             Project: Cassandra
>          Issue Type: Improvement
>            Reporter: Benjamin Roth
>
> I often encountered repair scenarios, where a lot of tiny repair streams were created. This results in hundrets, thousands or even ten-thousands super small SSTables (some bytes to some kbytes).
> This puts a lot of pressure on compaction and may even lead to a crash due to too many open files - I also encountered this.
> What could help to avoid this:
> After CASSANDRA-12888 is resolved, a tiny stream (e.g. < 100kb) could be sent through the write path to be buffered by memtables instead of creating an SSTable each.
> Without CASSANDRA-12888 this would break incremental repairs.



--
This message was sent by Atlassian JIRA
(v6.3.15#6346)