hadoop-common-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Jim Kellerman (JIRA)" <j...@apache.org>
Subject [jira] Commented: (HADOOP-2139) [hbase] Increase parallelism in region servers
Date Tue, 06 Nov 2007 23:06:56 GMT

    [ https://issues.apache.org/jira/browse/HADOOP-2139?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#action_12540617
] 

Jim Kellerman commented on HADOOP-2139:
---------------------------------------

Strategy:

Associate memcache with HStore instead of HRegion - reduces contention between most other
ops and scanners.

If a scanner is open on a memcache, and an update arrives, push the memcache on a stack and
create a new memcache for the update. This is unlikely to cause the generation of a lot of
memcaches since scanners don't get created nearly as often.

All operations other than Split lock the region for read. Locks for portions of a region will
be contended at the column level.

Splits need to lock the region for write since this is the only operation that cannot proceed
in parallel with others.


> [hbase] Increase parallelism in region servers
> ----------------------------------------------
>
>                 Key: HADOOP-2139
>                 URL: https://issues.apache.org/jira/browse/HADOOP-2139
>             Project: Hadoop
>          Issue Type: Improvement
>          Components: contrib/hbase
>    Affects Versions: 0.16.0
>            Reporter: Jim Kellerman
>            Assignee: Jim Kellerman
>             Fix For: 0.16.0
>
>
> There are a number of paths in the region server which block against one another including:
> - log rolling
> - cache flushes
> - region splitting
> - updates
> - scanners
> Investigate which can proceed in parallel and mechanisms for making some operations that
currently do not run in parallel.

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


Mime
View raw message