[
https://issues.apache.org/jira/browse/HBASE-5332?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13200233#comment-13200233
]
Nicolas Spiegelberg commented on HBASE-5332:
--------------------------------------------
The current proposal is to seed the random number generator with existing
persistent data:
{code}
java.util.Random(hash(Region Name + oldest file timestamp)).nextDouble()
{code}
Note that that the currently algorithm is normally an annoyance in 92 and above
because major compactions will normally go in the large compaction queue.
However, 90 and users who have the large compaction queue incorrectly
configured could experience high latency after restart because a lot of major
compactions will pile up.
> Deterministic Compaction Jitter
> -------------------------------
>
> Key: HBASE-5332
> URL: https://issues.apache.org/jira/browse/HBASE-5332
> Project: HBase
> Issue Type: Improvement
> Reporter: Nicolas Spiegelberg
> Assignee: Nicolas Spiegelberg
> Priority: Minor
>
> Currently, we add jitter to a compaction using "delay + jitter*(1 -
> 2*Math.random())". Since this is non-deterministic, we can get major
> compaction storms on server restart as half the Stores that were set to
> "delay + jitter" will now be set to "delay - jitter". We need a more
> deterministic way to jitter major compactions so this information can persist
> across server restarts.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators:
https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
