[jira] Commented: (HBASE-2832) Priorities and multi-threading for MemStore flushing

["Nicolas Spiegelberg (JIRA)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/HBASE-2832?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12980377#action_12980377 ] 

Nicolas Spiegelberg commented on HBASE-2832:
--------------------------------------------

Not sure about the state of this JIRA (Review Board seems to be down).  I believe the current design is to have thread-pools for compactions and to have one thread dedicated to kick in when we hit the block storefiles limit.  One idea thrown around internally is having a dedicated compaction thread with a lowered 'hbase.hstore.compaction.max.size' that will kick in earlier and only compact small files.  The idea is to keep the storefile count low for reads in the face of a couple long compactions (e.g. major compaction storm).

> Priorities and multi-threading for MemStore flushing
> ----------------------------------------------------
>
>                 Key: HBASE-2832
>                 URL: https://issues.apache.org/jira/browse/HBASE-2832
>             Project: HBase
>          Issue Type: New Feature
>          Components: regionserver
>            Reporter: Jonathan Gray
>            Assignee: Jonathan Gray
>            Priority: Critical
>             Fix For: 0.92.0
>
>
> Similar to HBASE-1476 and HBASE-2646 which are for compactions, but do this for flushes.
> Flushing when we hit the normal flush size is a low priority flush.  Other types of flushes (heap pressure, blocking client requests, etc) are high priority.
> Should have a tunable number of concurrent flushes.
> Will use the {{HBaseExecutorService}} and {{HBaseEventHandler}} introduced from master/zk changes.

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.