Return-Path: X-Original-To: apmail-hbase-issues-archive@www.apache.org Delivered-To: apmail-hbase-issues-archive@www.apache.org Received: from mail.apache.org (hermes.apache.org [140.211.11.3]) by minotaur.apache.org (Postfix) with SMTP id 927839585 for ; Wed, 30 May 2012 08:23:26 +0000 (UTC) Received: (qmail 25603 invoked by uid 500); 30 May 2012 08:23:26 -0000 Delivered-To: apmail-hbase-issues-archive@hbase.apache.org Received: (qmail 25482 invoked by uid 500); 30 May 2012 08:23:24 -0000 Mailing-List: contact issues-help@hbase.apache.org; run by ezmlm Precedence: bulk List-Help: List-Unsubscribe: List-Post: List-Id: Delivered-To: mailing list issues@hbase.apache.org Received: (qmail 25443 invoked by uid 99); 30 May 2012 08:23:23 -0000 Received: from issues-vm.apache.org (HELO issues-vm) (140.211.11.160) by apache.org (qpsmtpd/0.29) with ESMTP; Wed, 30 May 2012 08:23:23 +0000 Received: from isssues-vm.apache.org (localhost [127.0.0.1]) by issues-vm (Postfix) with ESMTP id 2C2F9142858 for ; Wed, 30 May 2012 08:23:23 +0000 (UTC) Date: Wed, 30 May 2012 08:23:23 +0000 (UTC) From: "chunhui shen (JIRA)" To: issues@hbase.apache.org Message-ID: <714483862.15241.1338366203182.JavaMail.jiratomcat@issues-vm> In-Reply-To: <1215661311.14885.1338359723326.JavaMail.jiratomcat@issues-vm> Subject: [jira] [Commented] (HBASE-6134) Speed up distribued split log MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: quoted-printable X-JIRA-FingerPrint: 30527f35849b9dde25b450d4833f0394 [ https://issues.apache.org/jira/browse/HBASE-6134?page=3Dcom.atlassian= .jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=3D1328= 5496#comment-13285496 ]=20 chunhui shen commented on HBASE-6134: ------------------------------------- @ted In my opition, we could do the improvement for split-worker. In the patch , I change the split logic which executed by the split-worker = same as local-master-split. It means, there are multi threads to creater writer and append hlog entry. Also, we found closing writer is quite slow, so make the closeStreams() exe= cuted by multi threads. With the patch, we test the time of distribued split log again, and reduce = the time from 165s+ to 38s+ (34hlog files 4 regionervers) =20 > Speed up distribued split log > ----------------------------- > > Key: HBASE-6134 > URL: https://issues.apache.org/jira/browse/HBASE-6134 > Project: HBase > Issue Type: Improvement > Components: wal > Reporter: chunhui shen > Assignee: chunhui shen > Fix For: 0.96.0 > > Attachments: HBASE-6134.patch > > > First=EF=BC=8Cwe do the test between local-master-split and distributed s= plit log > Environment=EF=BC=9A34 hlog files, 5 regionservers,(after kill one, only = 4 rs do ths splitting work) > local-master-split:60s+ > distributed-split-log:165s+ > In fact, in our production environment, distributed-split-log also took 6= 0s with 30 regionservers for 34 hlog files (regionserver may be in high loa= d) > We found split-worker split one log file took about 20s. > I think we should do the improvement for this. -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrato= rs: https://issues.apache.org/jira/secure/ContactAdministrators!default.jsp= a For more information on JIRA, see: http://www.atlassian.com/software/jira