You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@hbase.apache.org by "Mingjie Lai (JIRA)" <ji...@apache.org> on 2010/05/13 00:05:43 UTC
[jira] Commented: (HBASE-2468) Improvements to prewarm META cache
on clients
[ https://issues.apache.org/jira/browse/HBASE-2468?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12866758#action_12866758 ]
Mingjie Lai commented on HBASE-2468:
------------------------------------
I'm looking at this issue right now, and want to provide a patch it as a warmup.
My plan is to add a new HConnectionManager::locateRegionInMeta() which is dedicated for .META. table. It keeps the features of the original version (locate region for a table+row pair), in addition it puts all this table's region info to local cache by usig MetaScanner.
> Improvements to prewarm META cache on clients
> ---------------------------------------------
>
> Key: HBASE-2468
> URL: https://issues.apache.org/jira/browse/HBASE-2468
> Project: Hadoop HBase
> Issue Type: Improvement
> Components: client
> Reporter: Todd Lipcon
> Assignee: Mingjie Lai
> Fix For: 0.20.5
>
>
> A couple different use cases cause storms of reads to META during startup. For example, a large MR job will cause each map task to hit meta since it starts with an empty cache.
> A couple possible improvements have been proposed:
> - MR jobs could ship a copy of META for the table in the DistributedCache
> - Clients could prewarm cache by doing a large scan of all the meta for the table instead of random reads for each miss
> - Each miss could fetch ahead some number of rows in META
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.