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 49C9619652 for ; Wed, 13 Apr 2016 21:06:26 +0000 (UTC) Received: (qmail 58409 invoked by uid 500); 13 Apr 2016 21:06:26 -0000 Delivered-To: apmail-hbase-issues-archive@hbase.apache.org Received: (qmail 58362 invoked by uid 500); 13 Apr 2016 21:06:26 -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 58309 invoked by uid 99); 13 Apr 2016 21:06:25 -0000 Received: from arcas.apache.org (HELO arcas) (140.211.11.28) by apache.org (qpsmtpd/0.29) with ESMTP; Wed, 13 Apr 2016 21:06:25 +0000 Received: from arcas.apache.org (localhost [127.0.0.1]) by arcas (Postfix) with ESMTP id B17D42C1F6A for ; Wed, 13 Apr 2016 21:06:25 +0000 (UTC) Date: Wed, 13 Apr 2016 21:06:25 +0000 (UTC) From: "stack (JIRA)" To: issues@hbase.apache.org Message-ID: In-Reply-To: References: Subject: [jira] [Updated] (HBASE-12148) Remove TimeRangeTracker as point of contention when many threads writing a Store MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: 7bit X-JIRA-FingerPrint: 30527f35849b9dde25b450d4833f0394 [ https://issues.apache.org/jira/browse/HBASE-12148?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] stack updated HBASE-12148: -------------------------- Attachment: Screen Shot 2016-04-13 at 1.49.30 PM.png Screen Shot 2016-04-13 at 2.02.22 PM.png +1 on this last patch. I like the AtomicReference trick. Nice. I ran into this issue again while profiling. The TimeRangeTracker is reported as the location where we contend most in a ycsb workloadc randomread test doing about 300k ops a second. Attached are flight recorder before and after where TimeRangeTracker goes from being most contended item to not showing at all. Let me reattach patch and get a clean build then will apply everywhere. > Remove TimeRangeTracker as point of contention when many threads writing a Store > -------------------------------------------------------------------------------- > > Key: HBASE-12148 > URL: https://issues.apache.org/jira/browse/HBASE-12148 > Project: HBase > Issue Type: Sub-task > Components: Performance > Affects Versions: 2.0.0, 0.99.1 > Reporter: stack > Assignee: Walter Koetke > Fix For: 2.0.0, 1.3.0, 0.98.19 > > Attachments: 0001-In-AtomicUtils-change-updateMin-and-updateMax-to-ret.patch, 12148.addendum.txt, 12148.txt, 12148.txt, 12148v2.txt, 12148v2.txt, HBASE-12148-V3.patch, HBASE-12148-V3.patch, HBASE-12148.txt, HBASE-12148V2.txt, Screen Shot 2014-10-01 at 3.39.46 PM.png, Screen Shot 2014-10-01 at 3.41.07 PM.png, Screen Shot 2016-04-13 at 1.49.30 PM.png, Screen Shot 2016-04-13 at 2.02.22 PM.png, TimeRangeTracker.tiff > > -- This message was sent by Atlassian JIRA (v6.3.4#6332)