hadoop-hdfs-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Jean-Daniel Cryans (JIRA)" <j...@apache.org>
Subject [jira] Commented: (HDFS-918) Use single Selector and small thread pool to replace many instances of BlockSender for reads
Date Thu, 29 Jul 2010 20:28:20 GMT

    [ https://issues.apache.org/jira/browse/HDFS-918?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12893807#action_12893807
] 

Jean-Daniel Cryans commented on HDFS-918:
-----------------------------------------

I tested the latest full patch (not the pool one) and didn't see any issue, although it was
on a single machine and with block cache ON (but lots of evictions). I need to do some more
proper testing on my test cluster.

bq. Would people be interested in another JIRA focusing on just the pooling improvements

I am, and I think that having smaller patches will help getting this committed.

> Use single Selector and small thread pool to replace many instances of BlockSender for
reads
> --------------------------------------------------------------------------------------------
>
>                 Key: HDFS-918
>                 URL: https://issues.apache.org/jira/browse/HDFS-918
>             Project: Hadoop HDFS
>          Issue Type: Improvement
>          Components: data-node
>            Reporter: Jay Booth
>             Fix For: 0.22.0
>
>         Attachments: hbase-hdfs-benchmarks.ods, hdfs-918-20100201.patch, hdfs-918-20100203.patch,
hdfs-918-20100211.patch, hdfs-918-20100228.patch, hdfs-918-20100309.patch, hdfs-918-branch20-append.patch,
hdfs-918-branch20.2.patch, hdfs-918-pool.patch, hdfs-918-TRUNK.patch, hdfs-multiplex.patch
>
>
> Currently, on read requests, the DataXCeiver server allocates a new thread per request,
which must allocate its own buffers and leads to higher-than-optimal CPU and memory usage
by the sending threads.  If we had a single selector and a small threadpool to multiplex request
packets, we could theoretically achieve higher performance while taking up fewer resources
and leaving more CPU on datanodes available for mapred, hbase or whatever.  This can be done
without changing any wire protocols.

-- 
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