Return-Path: X-Original-To: apmail-cassandra-commits-archive@www.apache.org Delivered-To: apmail-cassandra-commits-archive@www.apache.org Received: from mail.apache.org (hermes.apache.org [140.211.11.3]) by minotaur.apache.org (Postfix) with SMTP id 3A37CF77C for ; Tue, 30 Apr 2013 07:52:26 +0000 (UTC) Received: (qmail 57146 invoked by uid 500); 30 Apr 2013 07:52:25 -0000 Delivered-To: apmail-cassandra-commits-archive@cassandra.apache.org Received: (qmail 56785 invoked by uid 500); 30 Apr 2013 07:52:17 -0000 Mailing-List: contact commits-help@cassandra.apache.org; run by ezmlm Precedence: bulk List-Help: List-Unsubscribe: List-Post: List-Id: Reply-To: dev@cassandra.apache.org Delivered-To: mailing list commits@cassandra.apache.org Received: (qmail 56722 invoked by uid 99); 30 Apr 2013 07:52:16 -0000 Received: from arcas.apache.org (HELO arcas.apache.org) (140.211.11.28) by apache.org (qpsmtpd/0.29) with ESMTP; Tue, 30 Apr 2013 07:52:16 +0000 Date: Tue, 30 Apr 2013 07:52:15 +0000 (UTC) From: "Vijay (JIRA)" To: commits@cassandra.apache.org Message-ID: In-Reply-To: References: Subject: [jira] [Commented] (CASSANDRA-5521) move IndexSummary off heap 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/CASSANDRA-5521?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13645365#comment-13645365 ] Vijay commented on CASSANDRA-5521: ---------------------------------- Pushed an update to https://github.com/Vijay2win/cassandra/commits/5521_v2 This update doesn't use int[] which v1 used, v2 uses kind of offheap index over index summary which is stored at the beginning of the memory block... Moves index summary completely offheap... :) I was not able to see any difference in performance between trunk and v2 using stress tool. Micro benchmark shows v2 is 6 seconds slower for 20 Bilion get's hence it is was not noticeable in stress tool, but the real benefit is less pauses and more data a node can hold... > move IndexSummary off heap > -------------------------- > > Key: CASSANDRA-5521 > URL: https://issues.apache.org/jira/browse/CASSANDRA-5521 > Project: Cassandra > Issue Type: Bug > Components: Core > Reporter: Jonathan Ellis > Assignee: Vijay > Fix For: 2.0 > > > IndexSummary can still use a lot of heap for narrow-row sstables. (It can also contribute to memory fragmentation because of the large arrays it creates.) -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators For more information on JIRA, see: http://www.atlassian.com/software/jira