Mailing-List: contact issues-help@hbase.apache.org; run by ezmlm
Precedence: bulk
Message-ID: <5680772.396701282062978196.JavaMail.jira@thor>
Date: Tue, 17 Aug 2010 12:36:18 -0400 (EDT)
From: "Jean-Daniel Cryans (JIRA)" <jira@apache.org>
To: issues@hbase.apache.org
Subject: [jira] Commented: (HBASE-2467) Concurrent flushers in HLog sync
 using HDFS-895
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: 7bit


    [ https://issues.apache.org/jira/browse/HBASE-2467?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12899459#action_12899459 ] 

Jean-Daniel Cryans commented on HBASE-2467:
-------------------------------------------

bq. What did you do to the patch J-D?

Mostly just a refresh, and I removed an now-unused boolean on sync().

bq.  In the hadoop we ship, we should include hdfs-895?

I think it would be great.

bq. What should we do about group commit. Open a new issue?

We don't need it anymore, but having deferred log flush could still be useful. Should be in the same issue.

bq. If hdfs-895 is not in place, we just run slower, is that right?

Yes, under concurrent writers it's much slower without 895.

bq. If concurrency, should 'private boolean logRollRequested' be volatile?

As a matter of fact, it's under 2 different sync (this and updateLock). Maybe safer to put volatile?

bq. Why not do the syncTime and syncOps under the updateLock rather than do a synchronize on 'this'?

Makes sense.

> 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
>             Fix For: 0.90.0
>
>         Attachments: HBASE-2467-v2.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.