hadoop-hdfs-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Todd Lipcon (Commented) (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (HDFS-3023) Optimize entries in edits log for persistBlocks calls
Date Tue, 28 Feb 2012 17:29:46 GMT

    [ https://issues.apache.org/jira/browse/HDFS-3023?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13218375#comment-13218375

Todd Lipcon commented on HDFS-3023:

I do think the log size is the major difference causing the performance issues in HDFS-3010.
In my initial benchmarks last night (which included only one of the above optimizations) the
regression was only 7% vs the ~20% I had seen in earlier testing. But, these results aren't
reliable since I also switched to a different cluster between benchmarks. I'll rerun the benchmarks

Also worth noting - you referred to "option #2 vs option #3" above, but really option #3 is
combined with option #2 for extra savings (ie the OP_UPDATE_BLOCKS is around half to 1/3 the
size because of the delta-vint-encoded genstamps and sizes)
> Optimize entries in edits log for persistBlocks calls
> -----------------------------------------------------
>                 Key: HDFS-3023
>                 URL: https://issues.apache.org/jira/browse/HDFS-3023
>             Project: Hadoop HDFS
>          Issue Type: Improvement
>          Components: name-node, performance
>    Affects Versions: HA branch (HDFS-1623), 0.23.2
>            Reporter: Todd Lipcon
>            Assignee: Todd Lipcon
>         Attachments: hdfs-3023-HDFS-1623.txt
> One of the performance issues noticed in the HA branch is due to the much larger edit
logs, now that we are writing OP_ADD transactions to the edit log on every block allocation.
We can condense these calls down in two ways:
> 1) use variable-length integers for the block list length, size, and genstamp (most of
these end up fitting in far less than 8 bytes)
> 2) use delta-coding for the genstamp and block size for any blocks after the first block
(most blocks will be the same size and only slightly higher genstamps)
> 3) introduce a new OP_UPDATE_BLOCKS transaction that doesn't re-serialize metadata information
like lease owner, permissions, etc
> 4) allow OP_UPDATE_BLOCKS to only re-serialize the blocks that have changed for a given

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


View raw message