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 74180107C5 for ; Fri, 24 Jan 2014 14:08:26 +0000 (UTC) Received: (qmail 31757 invoked by uid 500); 24 Jan 2014 14:08:23 -0000 Delivered-To: apmail-cassandra-user-archive@cassandra.apache.org Received: (qmail 31500 invoked by uid 500); 24 Jan 2014 14:08:22 -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 31486 invoked by uid 99); 24 Jan 2014 14:08:20 -0000 Received: from nike.apache.org (HELO nike.apache.org) (192.87.106.230) by apache.org (qpsmtpd/0.29) with ESMTP; Fri, 24 Jan 2014 14:08:20 +0000 X-ASF-Spam-Status: No, hits=2.2 required=5.0 tests=HTML_MESSAGE,SPF_PASS X-Spam-Check-By: apache.org Received-SPF: pass (nike.apache.org: domain of prvs=09460c614=john.anderstedt@svenskaspel.se designates 78.108.6.27 as permitted sender) Received: from [78.108.6.27] (HELO smtp2.svenskaspel.se) (78.108.6.27) by apache.org (qpsmtpd/0.29) with ESMTP; Fri, 24 Jan 2014 14:08:14 +0000 DKIM-Signature: v=1; a=rsa-sha256; c=simple/simple; d=svenskaspel.se; i=@svenskaspel.se; q=dns/txt; s=dkim; t=1390572495; x=1422108495; h=from:to:date:subject:message-id:references:in-reply-to: mime-version; bh=zLjKuhvyEArsOeosh9smQp9mRlfStsAyhMOK/a56ftk=; b=DyB+617BTTdts6aNNLdY4mGP68eCFiWtsdRFqLXlUyXVak1inijsKrHr c+oxD7no2amMAqTRRTWBHtwqvMUGx5gfu4f/jbRg0pVrZUQn8OvGEx1kr o9kkwGeJ0eZ63L+n2irVjWIRDw3xH0M7IE2tq4ujB10Ajk2d3bCi2W4dE pelmIKRqM0GNttksNvjFzknX4GS4DyntOsQT3fc/9GrZPOD60fKi1TFWU s+W3Eh1bNsbjTMCVLEk6fQ2JXw9iibkkIVkIWpQ9+CMesT/8aRLY+CMid 7PqMEct+yhbOIA80E8Z44OO7pFmkZdb3C9ua80hYSPbXMahMBUWaAqP+k A==; X-IronPort-AV: E=Sophos;i="4.95,712,1384297200"; d="scan'208,217";a="2282059" Received: from rsbgchs01.sbg.spel.se (HELO rsbgchs01.ad.spel.se) ([172.23.1.94]) by smtp2.vby.svenskaspel.se with ESMTP/TLS/RC4-MD5; 24 Jan 2014 15:07:55 +0100 Received: from RSBGMBS01.ad.spel.se ([172.23.1.91]) by rsbgchs01.ad.spel.se ([172.23.1.94]) with mapi; Fri, 24 Jan 2014 15:07:53 +0100 From: John Anderstedt To: "user@cassandra.apache.org" Date: Fri, 24 Jan 2014 15:07:53 +0100 Subject: Re: Cassandra Performance Testing Thread-Topic: Cassandra Performance Testing Thread-Index: Ac8ZDayaBOz7OeO0SEaXbOE/jrMo2g== Message-ID: References: In-Reply-To: Accept-Language: sv-SE Content-Language: en-US X-MS-Has-Attach: X-MS-TNEF-Correlator: acceptlanguage: sv-SE Content-Type: multipart/alternative; boundary="_000_CC9C1E223400459D938443D262503EDDsvenskaspelse_" MIME-Version: 1.0 X-Virus-Checked: Checked by ClamAV on apache.org --_000_CC9C1E223400459D938443D262503EDDsvenskaspelse_ MIME-Version: 1.0 Content-Type: text/plain; charset="windows-1252" Content-Transfer-Encoding: quoted-printable It sounds to me that the limitation in this setup is the disks. if it=92s in a mirror the cost for write=92s is the dubble. If you have the flatfile and the db on the same disk there will be a lot of= io wait. There is also a question of diskspace and fragmentation, if the flat file o= ccupies 1,2TB of a total of 2TB and then add the db to that. It will fill u= p and on the way it would go slower by the hour because the background proc= esses that runs the compaction needs space to run. make sense? mvh / regards John 24 jan 2014 kl. 13:46 skrev Devin Pinkston >: Hello, I am using a single node Cassandra setup with version 2.0.4 to do some simp= le performance testing. I generated a 1.2TB flat file from DBGEN (TPC-H), = and I am loading that into Cassandra. I used the =93COPY FROM=94 method fr= om the CQLSH. My question/problem, the import has been running for over two days! Is the= re something I am potentially doing wrong? I setup the single node Cassand= ra from the getting started page. Specs for the server being used, (Ubuntu 12.0.4.3 LTS x64) Model 1950 Dell PowerEdge 1950 Server II Processor 2 2x Intel Quad Core 2.33GHz E5345 8MB Memory Installed 32GB 32GB PC2-5300F Fully Buffered Memory Memory Size 4GB Eight Slots Available: 8 x 4GB Memory Sticks Hard Drives Included 2x 1TB 7.2K SATA Hard Drives Thanks The information contained in this transmission may contain privileged and c= onfidential information. It is intended only for the use of the person(s) named above. If you are not the intended recipient, you are hereby notified that any rev= iew, dissemination, distribution or duplication of this communication is st= rictly prohibited. If you are not the intended recipient, please contact the sender by reply e= -mail and destroy all copies of the original message. Technica Corporation does not represent this e-mail to be free from any vir= us, fault or defect and it is therefore the responsibility of the recipient= to first scan it for viruses, faults and defects. To reply to our e-mail administrator directly, please send an e-mail to pos= tmaster@technicacorp.com. Thank you. AB SVENSKA SPEL 621 80 Visby Norra Hansegatan 17, Visby V=E4xel: +4610-120 00 00 https://svenskaspel.se Please consider the environment before printing this email --_000_CC9C1E223400459D938443D262503EDDsvenskaspelse_ MIME-Version: 1.0 Content-Type: text/html; charset="windows-1252" Content-Transfer-Encoding: quoted-printable It sounds to me that t= he limitation in this setup is the disks.
if it=92s in a mirror the= cost for write=92s is the dubble. 

If you ha= ve the flatfile and the db on the same disk there will be a lot of io wait.=

There is also a question of diskspace and fragmen= tation, if the flat file occupies 1,2TB of a total of 2TB and then add the = db to that. It will fill up and on the way it would go slower by the hour b= ecause the background processes that runs the compaction needs space to run= .

make sense?

mvh / regar= ds
John

24 jan 2014 kl. 13:46 skrev Devin= Pinkston <dpinkston@techn= icacorp.com>:

Hello,
 
I am using a single node Cassandra setup with version 2.= 0.4 to do some simple performance testing.  I generated a 1.2TB flat f= ile from DBGEN (TPC-H), and I am loading that into Cassandra.  I used = the =93COPY FROM=94 method from the CQLSH.
 
My question/problem, the import ha= s been running for over two days!  Is there something I am potentially= doing wrong?  I setup the single node Cassandra from the getting star= ted page. 
 
Specs for the server being used, (Ubuntu 12.0.4.3 LTS x64)
 
Model
1950
= Dell Power= Edge 1950 Server II
Processor
2
2x Intel Quad Core 2.33GHz E5345 8MB
Memory Installed
32G= B
32GB PC2-5300F Fully = Buffered Memory
Memory Size
4GB
Eight Slots Available: 8 x 4= GB Memory Sticks 
Hard Drives
Included
<= div style=3D"margin: 0in 0in 0.0001pt; font-size: 11pt; font-family: Calibr= i, sans-serif;">2x 1TB 7.2K  SATA Hard Drives
 
 
Thanks
<= /div>

The information contained in this transmission may contain privileg= ed and confidential information. = ;
It is intended only for the use of the person(s) named above. 
If you are not the int= ended recipient, you are hereby notified that any review, dissemination, di= stribution or duplication of this communication is strictly prohibited. 
If you are not the inten= ded recipient, please contact the sender by reply e-mail and destroy all co= pies of the original message. 
Technica Corporation does not represent this e-mail to be free fro= m any virus, fault or defect and it is therefore the responsibility of the = recipient to first scan it for viruses, faults and defects. 
To reply to our e-mail administrator= directly, please send an e-mail to&n= bsp;postmaster@technicacorp.com. Thank = you.


A= B SVENSKA SPEL
621 80 Visby
Norra Hansegatan 17, Visby
V=E4xel: +4= 610-120 00 00
https://svenskaspel.se

P