[jira] Commented: (HBASE-2467) Concurrent flushers in HLog sync using HDFS-895

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

    [ https://issues.apache.org/jira/browse/HBASE-2467?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12931176#action_12931176 ] 

Nicolas Spiegelberg commented on HBASE-2467:
--------------------------------------------

How much operational experience have you had with production machines running HDFS-895 + HBASE-2467?  I did a code review of HDFS-895 and feel like it's pretty solid.  We have it applied to our internal branch, but we don't have HBASE=2467 applied.  Although I think the HDFS-895 code is good, there's no question that it's very intricate and needs plenty of test time with large traffic before we should rely on it.  I hear some of you guys have been running it on production traffic for months, is that correct?  If so, I think that's sufficient test time and we can commit this.

> Concurrent flushers in HLog sync using HDFS-895
> -----------------------------------------------
>
>                 Key: HBASE-2467
>                 URL: https://issues.apache.org/jira/browse/HBASE-2467
>             Project: HBase
>          Issue Type: Improvement
>          Components: regionserver
>            Reporter: Todd Lipcon
>            Assignee: Todd Lipcon
>            Priority: Blocker
>             Fix For: 0.90.0
>
>         Attachments: HBASE-2467-v2.patch, HBASE-2467-v3.patch, HBASE-2467-v4.patch, hbase-2467.txt
>
>
> HDFS-895 changes hflush() to be able to run concurrently from multiple threads, where flushes can be concurrent with further writes to the same file.
> We need to rip out/amend the group commit code a bit to take advantage of this.

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