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 108114154 for ; Wed, 15 Jun 2011 19:44:22 +0000 (UTC) Received: (qmail 23379 invoked by uid 500); 15 Jun 2011 19:44:19 -0000 Delivered-To: apmail-cassandra-user-archive@cassandra.apache.org Received: (qmail 23350 invoked by uid 500); 15 Jun 2011 19:44:19 -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 23342 invoked by uid 99); 15 Jun 2011 19:44:19 -0000 Received: from nike.apache.org (HELO nike.apache.org) (192.87.106.230) by apache.org (qpsmtpd/0.29) with ESMTP; Wed, 15 Jun 2011 19:44:19 +0000 X-ASF-Spam-Status: No, hits=2.2 required=5.0 tests=FREEMAIL_FROM,HTML_MESSAGE,RCVD_IN_DNSWL_NONE,RFC_ABUSE_POST,SPF_PASS,T_TO_NO_BRKTS_FREEMAIL X-Spam-Check-By: apache.org Received-SPF: pass (nike.apache.org: local policy) Received: from [98.138.91.160] (HELO nm30-vm3.bullet.mail.ne1.yahoo.com) (98.138.91.160) by apache.org (qpsmtpd/0.29) with SMTP; Wed, 15 Jun 2011 19:44:10 +0000 Received: from [98.138.90.54] by nm30.bullet.mail.ne1.yahoo.com with NNFMP; 15 Jun 2011 19:43:49 -0000 Received: from [98.138.86.156] by tm7.bullet.mail.ne1.yahoo.com with NNFMP; 15 Jun 2011 19:43:49 -0000 Received: from [127.0.0.1] by omp1014.mail.ne1.yahoo.com with NNFMP; 15 Jun 2011 19:43:46 -0000 X-Yahoo-Newman-Property: ymail-3 X-Yahoo-Newman-Id: 903725.48417.bm@omp1014.mail.ne1.yahoo.com Received: (qmail 14549 invoked by uid 60001); 15 Jun 2011 19:43:46 -0000 DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=yahoo.com; s=s1024; t=1308167026; bh=xDimYvc5m+/JGVLHbHXp/O+gyKNqRpOOJK4JIAxTI8M=; h=Message-ID:X-YMail-OSG:Received:X-Mailer:References:Date:From:Subject:To:In-Reply-To:MIME-Version:Content-Type; b=wxTt+FFxHa4BNgSEoTFrjWI2rktKRrm0q2fScEpkOWBV6JsJ13dJQxbpa/F8QsBvk23KV8MhXnjV8gqOcAJz6YFC+sZzJQ/5lsKhVBl+5MkTY4ptYpwr+bUZiZELg2EE5nb1fZuWg0gMKLRqBoimpKAfGQFelkdHKJG96LZUliA= DomainKey-Signature: a=rsa-sha1; q=dns; c=nofws; s=s1024; d=yahoo.com; h=Message-ID:X-YMail-OSG:Received:X-Mailer:References:Date:From:Subject:To:In-Reply-To:MIME-Version:Content-Type; b=od94RpaCMP3cTI4YJvhaVGHtTQZTaNQSRg+XP1WQu69KF2eAF5wO7rJvr/PTnf5TSUDgEj7wWNBIm4mrcbaMpTC9MqCoe45w0fBKrg9h1RPG62yI7KyQbbPAjNC/38iSrd1r2+GeVN8jkFqmEA2l4LgC1f3+G1a4UrVuLJmDzOY=; Message-ID: <792031.7261.qm@web121510.mail.ne1.yahoo.com> X-YMail-OSG: vTmGnT0VM1n.stH8ZNc4sF9yQgaIKkou8YEdzYduz7CdbHa .fRqkpvuV7Ek0_6HfnKHl5.wy5GG8oYw.xS.KtjTLg8H1Uj1aeTOGsLzhIUU UW7aJIN1f0jA.atkdwPJBsM8zxu42gFngKqOnw7Jojge8PMCe6X811TX7NcE nxppV6psq.rytjME2BC_LsnhPbnEg1k38MBH7qI38OQySYt2.lP5Dl7UsS5n Xb87mn_6Lv4y6OggnUPEu2lF35xb8T75d9feTsbFyBGX0_Dc1hfz5WEc4Y.K L6HiTV1sCJA2xZ0FRr.P.2jURUYAteSzcMeQFL5IZkw4zUhXhM6.nDbrQRWe 30PhszyhEXAXOCStmCRV.ulCn.IYk_Yq6tCsPrAwOck0ArXAqIyFZyy2pRLR aKveTA8Y.0xcAO2NfeJOYGoiKPSNL99QjdP9Ev_NYxeFnJxv9jaLOwv4FLnP m.bVz2oR9lwrFMBE5AXJaa7MHTBtYmlOURguey_XsaQnyqm3jZKQ5HArxZto 78eDgx3Tml6pxxEfNHhjK3h1vvqGCMgpQB7Y- Received: from [63.85.50.241] by web121510.mail.ne1.yahoo.com via HTTP; Wed, 15 Jun 2011 12:43:46 PDT X-Mailer: YahooMailRC/570 YahooMailWebService/0.8.111.304355 References: <012801cc2af4$f9003930$eb00ab90$@gmail.com> <573619.55346.qm@web121520.mail.ne1.yahoo.com> Date: Wed, 15 Jun 2011 12:43:46 -0700 (PDT) From: Selva Kumar Subject: Re: Multi data center configuration - A question on read correction To: user@cassandra.apache.org In-Reply-To: MIME-Version: 1.0 Content-Type: multipart/alternative; boundary="0-196571177-1308167026=:7261" X-Virus-Checked: Checked by ClamAV on apache.org --0-196571177-1308167026=:7261 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: quoted-printable Thanks Jonathan. Can we turn off RR by READ_REPAIR_CHANCE.=3D 0. Please adv= ice.=0A=0ASelva=0A=0A=0A=0A=0A________________________________=0AFrom: Jona= than Ellis =0ATo: user@cassandra.apache.org=0ASent: Tue,= June 14, 2011 8:59:41 PM=0ASubject: Re: Multi data center configuration - = A question on read correction=0A=0AThat's just read repair sending MD5s of = the data for comparison.=A0 So=0Anet traffic is light.=0A=0AYou can turn of= f RR but the downsides can be large.=A0 Turning it down=0Ato say 10% can be= reasonable tho.=0A=0ABut again, if network traffic is your concern you sho= uld be fine.=0A=0AOn Tue, Jun 14, 2011 at 8:44 PM, Selva Kumar wrote:=0A> I have setup a multiple data center configuration in Ca= ssandra. My primary=0A> intention is to minimize the network traffic betwee= n DC1 and DC2. Want DC1=0A> read requests be served with out reaching DC2 n= odes. After going through=0A> documentation, i felt following setup would d= o.=0A>=0A>=0A> Replica Placement Strategy: NetworkTopologyStrategy=0A> Repl= ication Factor: 3=0A> strategy_options:=0A> DC1 : 2=0A> DC2 : 1=0A> endpoin= t_snitch: org.apache.cassandra.locator.PropertyFileSnitch=0A> Read Consiste= ncy Level: LOCAL_QUORUM=0A> Write Consistency Level: LOCAL_QUORUM=0A>=0A> F= ile: cassandra-topology.properties=0A> # Cassandra Node IP=3DData Center:Ra= ck=0A> 10.10.10.149=3DDC1:RAC1=0A> 10.10.10.150=3DDC1:RAC1=0A> 10.10.10.151= =3DDC1:RAC1=0A>=0A> 10.20.10.153=3DDC2:RAC1=0A> 10.20.10.154=3DDC2:RAC1=0A>= # default for unknown nodes=0A> default=3DDC1:RAC1=0A>=0A> Question I have= :=0A> 1. Created a java program to test. It was querying with consistency l= evel=0A> LOCAL_QUORUM on a DC1 node. Read count(Through cfstats) on the DC2= node=0A> showed read happened there too. Is it because of read correction?= . Is there=0A> way to avoid doing read correction in DC2 nodes, when we que= ry DC1 nodes.=0A>=0A> Thanks=0A> Selva=0A=0A=0A=0A-- =0AJonathan Ellis=0APr= oject Chair, Apache Cassandra=0Aco-founder of DataStax, the source for prof= essional Cassandra support=0Ahttp://www.datastax.com=0A --0-196571177-1308167026=:7261 Content-Type: text/html; charset=iso-8859-1 Content-Transfer-Encoding: quoted-printable
Thanks Jonathan. Can we turn off RR by READ_REPAIR_CHA= NCE.=3D 0. Please advice.
=0A
 
=0A
Selva
= =0A

=0A
=0A
=0AFrom: Jonathan Ellis <jbellis@gm= ail.com>
To: user@cas= sandra.apache.org
Sent: = Tue, June 14, 2011 8:59:41 PM
Subje= ct: Re: Multi data center configuration - A question on read cor= rection

That's just read repair sending MD5s of the data for = comparison.  So
net traffic is light.

You can turn off RR bu= t the downsides can be large.  Turning it down
to say 10% can be re= asonable tho.

But again, if network traffic is your concern you shou= ld be fine.

On Tue, Jun 14, 2011 at 8:44 PM, Selva Kumar <wwgsel= va@yahoo.com> wrote:
> I have setup a multiple data center con= figuration in Cassandra. My primary
> intention is to minimize the network traffic between DC1 and DC2. Want DC1
> read requests be ser= ved with out reaching DC2 nodes. After going through
> documentation,= i felt following setup would do.
>
>
> Replica Placement= Strategy: NetworkTopologyStrategy
> Replication Factor: 3
> st= rategy_options:
> DC1 : 2
> DC2 : 1
> endpoint_snitch: or= g.apache.cassandra.locator.PropertyFileSnitch
> Read Consistency Leve= l: LOCAL_QUORUM
> Write Consistency Level: LOCAL_QUORUM
>
&g= t; File: cassandra-topology.properties
> # Cassandra Node IP=3DData C= enter:Rack
> 10.10.10.149=3DDC1:RAC1
> 10.10.10.150=3DDC1:RAC1<= BR>> 10.10.10.151=3DDC1:RAC1
>
> 10.20.10.153=3DDC2:RAC1
= > 10.20.10.154=3DDC2:RAC1
> # default for unknown nodes
> de= fault=3DDC1:RAC1
>
> Question I have:
> 1. Created a java= program to test. It was querying with consistency level
> LOCAL_QUOR= UM on a DC1 node. Read count(Through cfstats) on the DC2 node
> showed read = happened there too. Is it because of read correction?. Is there
> way= to avoid doing read correction in DC2 nodes, when we query DC1 nodes.
&= gt;
> Thanks
> Selva



--
Jonathan Ellis
Pr= oject Chair, Apache Cassandra
co-founder of DataStax, the source for pro= fessional Cassandra support
http://www.datastax.com
--0-196571177-1308167026=:7261--