Return-Path: X-Original-To: apmail-cassandra-user-archive@www.apache.org Delivered-To: apmail-cassandra-user-archive@www.apache.org Received: from mail.apache.org (hermes.apache.org [140.211.11.3]) by minotaur.apache.org (Postfix) with SMTP id EE7487CC6 for ; Fri, 18 Nov 2011 00:54:11 +0000 (UTC) Received: (qmail 83796 invoked by uid 500); 18 Nov 2011 00:54:09 -0000 Delivered-To: apmail-cassandra-user-archive@cassandra.apache.org Received: (qmail 83770 invoked by uid 500); 18 Nov 2011 00:54:09 -0000 Mailing-List: contact user-help@cassandra.apache.org; run by ezmlm Precedence: bulk List-Help: List-Unsubscribe: List-Post: List-Id: Reply-To: user@cassandra.apache.org Delivered-To: mailing list user@cassandra.apache.org Received: (qmail 83762 invoked by uid 99); 18 Nov 2011 00:54:09 -0000 Received: from athena.apache.org (HELO athena.apache.org) (140.211.11.136) by apache.org (qpsmtpd/0.29) with ESMTP; Fri, 18 Nov 2011 00:54:09 +0000 X-ASF-Spam-Status: No, hits=-0.1 required=5.0 tests=HTML_MESSAGE,RCVD_IN_DNSWL_MED,SPF_PASS X-Spam-Check-By: apache.org Received-SPF: pass (athena.apache.org: domain of prvs=29630b25f=bburruss@expedia.com designates 216.251.112.221 as permitted sender) Received: from [216.251.112.221] (HELO mx1a.expedia.com) (216.251.112.221) by apache.org (qpsmtpd/0.29) with ESMTP; Fri, 18 Nov 2011 00:54:03 +0000 X-SRBS: None X-HAT: Sender Group RELAYLIST, Policy $RELAYED applied. Received: from unknown (HELO CHC-EXCHNA02.SEA.CORP.EXPECN.COM) ([10.184.69.26]) by mx1a.sea.corp.expecn.com with ESMTP; 17 Nov 2011 17:53:42 -0700 Received: from CHC-CCR02.SEA.CORP.EXPECN.COM ([fe80:0000:0000:0000:0000:0000:0.0.0.1]) by CHC-EXCHNA02.SEA.CORP.EXPECN.COM ([10.184.69.26]) with mapi; Thu, 17 Nov 2011 16:53:41 -0800 From: Todd Burruss To: "user@cassandra.apache.org" Date: Thu, 17 Nov 2011 16:53:40 -0800 Subject: ParNew and caching Thread-Topic: ParNew and caching Thread-Index: AcyljIPgACm7uaguQPagfCXJdxz6Wg== Message-ID: Accept-Language: en-US Content-Language: en-US X-MS-Has-Attach: X-MS-TNEF-Correlator: user-agent: Microsoft-MacOutlook/14.13.0.110805 acceptlanguage: en-US Content-Type: multipart/alternative; boundary="_000_CAEAEC9475B5bburrussexpediacom_" MIME-Version: 1.0 --_000_CAEAEC9475B5bburrussexpediacom_ Content-Type: text/plain; charset="Windows-1252" Content-Transfer-Encoding: quoted-printable I'm using cassandra 1.0. Been doing some testing on using cass's cache. W= hen I turn it on (using the CLI) I see ParNew jump from 3-4ms to 200-300ms.= This really screws with response times, which jump from ~25-30ms to 1300+= ms. I've increase new gen and that helps, but still this is suprising to m= e, especially since 1.0 defaults to the SerializingCacheProvider =96 off he= ap. The interesting tid bit is that I have wide rows. 70k+ columns per row, ~5= 0 bytes per column value. The cache only must be about 400 rows to catch a= ll the data per node and JMX is reporting 100% cache hits. Nodetool ring r= eports < 2gb per node, my heap is 6gb and total RAM is 16gb. Thoughts? --_000_CAEAEC9475B5bburrussexpediacom_ Content-Type: text/html; charset="Windows-1252" Content-Transfer-Encoding: quoted-printable

I'm using cassandra 1.0. = Been doing some testing on using cass's cache. When I turn it on (usi= ng the CLI) I see ParNew jump from 3-4ms to 200-300ms. This really sc= rews with response times, which jump from ~25-30ms to 1300+ms. I'= ve increase new gen and that helps, but still this is suprising to me, espe= cially since 1.0 defaults to the SerializingCacheProvider =96 off heap.

The interesting tid bit is that I have wide rows. &nb= sp;70k+ columns per row, ~50 bytes per column value. The cache on= ly must be about 400 rows to catch all the data per node and JMX is reporti= ng 100% cache hits. Nodetool ring reports < 2gb per node, my heap = is 6gb and total RAM is 16gb.

Thoughts?

--_000_CAEAEC9475B5bburrussexpediacom_--