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 17E7C10902 for ; Wed, 18 Sep 2013 15:32:12 +0000 (UTC) Received: (qmail 69837 invoked by uid 500); 18 Sep 2013 15:32:07 -0000 Delivered-To: apmail-cassandra-user-archive@cassandra.apache.org Received: (qmail 69661 invoked by uid 500); 18 Sep 2013 15:32:07 -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 69440 invoked by uid 99); 18 Sep 2013 15:32:05 -0000 Received: from nike.apache.org (HELO nike.apache.org) (192.87.106.230) by apache.org (qpsmtpd/0.29) with ESMTP; Wed, 18 Sep 2013 15:32:05 +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 post@fantasista.no designates 213.236.237.140 as permitted sender) Received: from [213.236.237.140] (HELO mx1.mailserveren.com) (213.236.237.140) by apache.org (qpsmtpd/0.29) with ESMTP; Wed, 18 Sep 2013 15:31:58 +0000 Received: from localhost ([127.0.0.1]) by mx1.mailserveren.com with esmtpa (Exim 4.80.1) (envelope-from ) id 1VMJj7-0002hq-Bj for user@cassandra.apache.org; Wed, 18 Sep 2013 17:31:37 +0200 Message-Id: <2c1c4b7ad519c5ccdb7299177a523bd26539c269@pop3.fantasista.no> From: "Vegard Berget" Reply-To: "Vegard Berget" To: user@cassandra.apache.org X-Mailer: Atmail 6.6.2.11727 X-Originating-IP: 46.19.16.3 in-reply-to: Subject: =?UTF-8?B?UmU6IFdoeSBkb24ndCB5b3Ugc3RhcnQgb2ZmIHdpdGggYSDigJxzaW5nbGUgJiBzbWFsbOKAnSBDYXNzYW5kcmEgc2VydmVyIGFzIHlvdSB1c3VhbGx5IGRvIGl0IHdpdGggTXlTUUw/?= Date: Wed, 18 Sep 2013 17:31:37 +0200 Content-Type: multipart/alternative; boundary="=_d491296a1c250e59f5b6d14ae1ac923a" MIME-Version: 1.0 X-Virus-Checked: Checked by ClamAV on apache.org --=_d491296a1c250e59f5b6d14ae1ac923a Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: quoted-printable Hi,=0AThe idea behind Cassandra is not the same as for MySQL. =C2=A0Firs= t of all=0Ayou can't get fault tolerance with one node. =C2=A0I don't th= ink Cassandra=0Anodes are more prone to be unavailable, but by using rep= lication you=0Acan get more availability right away. =C2=A0If you have m= ultiple instances=0Awhen you start out, you have the right configuration= when things get=0Atoo large for one node. =C2=A0Then there is the issue= that Cassandra is=0Aoften used with very large data or very high traffi= c, to which it=0Ascales very well. =C2=A0If you don't expect to ever sca= le beyond one node=0A(because traffic will never be high, you will never= have much data or=0Ayou never need the extra availability) then you of= course could use=0Aone single node without lots of RAM, but it _could_= also be that *SQL=0Ais a better option? =C2=A0=0AIf you are a startup w= ith great plans but no money and not much=0Adata/traffic (yet:), or just= want to test Cassandra, then I still=0Awould suggest >1 nodes (3?), and= instead reduce RAM and/or CPU to get=0Athe right price. =C2=A0In that c= ase you could try to use =C2=A0Tomcat on one=0Aof the nodes. =C2=A0This= way you can see and test what Cassandra does=0Areally well. =C2=A0It is= not really that Cassandra demands more than other=0Adatabases, but more= that the problems it is designed to solve demands=0Amore than some othe= r problems. =C2=A0FYI, I use 2-3 nodes with 2gb RAM=0Aeach for testing,= and have never had any problems with that. =C2=A0=0A.vegard,=0A----- Or= iginal Message -----=0AFrom: user@cassandra.apache.org=0ATo:"user" =0ACc= :=0ASent:Wed, 18 Sep 2013 19:04:27 +0530=0ASubject:Why don't you start o= ff with a =E2=80=9Csingle & small=E2=80=9D Cassandra=0Aserver as you usu= ally do it with MySQL?=0A=0A=09 For any website just starting out, the l= oad initially is minimal &=0Agrows with a slow pace initially. People us= ually start with their=0AMySQL based sites with a single server(***that= too a VPS not a=0Adedicated server) running as both app server as well= as DB server &=0Ausually get too far with this setup & only as they fee= l the need they=0Aseparate the DB from the app server giving it a separa= te VPS server.=0AThis is what a start up expects the things to be while= planning about=0Aresources procurement. =0A=0A=09 But so far what I hav= e seen, it's something very different with=0ACassandra. People usually r= ecommend starting out with atleast a 3 node=0Acluster, (on dedicated ser= vers) with lots & lots of RAM. 4GB or 8GB=0ARAM is what they suggest to= start with. So is it that Cassandra=0Arequires more hardware resources= in comparison to MySQL, for a website=0Ato deliver similar performance,= serve similar load/ traffic & same=0Aamount of data. I understand about= higher storage requirements of=0ACassandra due to replication but what= about other hardware resources ?=0A=0A=0A=09 Can't we start off with Ca= ssandra based apps just like MySQL.=0AStarting with 1 or 2 VPS & adding= more whenever there's a need ? =0A=0A=09 I don't want to compare apples= with oranges. I just want to know how=0Amuch more dangerous situation I= may be in when I start out with a=0Asingle node VPS based cassandra ins= tallation Vs a single node VPS=0Abased MySQL installation. Difference be= tween these two situations. Are=0Acassandra servers more prone to be una= vailable than MySQL servers ?=0AWhat is bad if I put tomcat too along wi= th Cassandra as people use=0ALAMP stack on single server. =0A=0A=09 ----= ----- =0A=0A=09 _This question is also posted at StackOverflow here [1]= =C2=A0& has an=0Aopen bounty worth +50 rep._ =0A=0ALinks:=0A------=0A[1= ]=0Ahttp://stackoverflow.com/questions/18462530/why-dont-you-start-off-w= ith-a-single-small-cassandra-server-as-you-usually=0A --=_d491296a1c250e59f5b6d14ae1ac923a Content-Type: text/html; charset=UTF-8 Content-Transfer-Encoding: quoted-printable Hi,

The idea behind Cassan= dra is not the same as for MySQL. =C2=A0First of all you can't get fault= tolerance with one node. =C2=A0I don't think Cassandra nodes are more p= rone to be unavailable, but by using replication you can get more availa= bility right away. =C2=A0If you have multiple instances when you start o= ut, you have the right configuration when things get too large for one n= ode. =C2=A0
Then there is the issue that Cassandra is often us= ed with very large data or very high traffic, to which it scales very we= ll. =C2=A0If you don't expect to ever scale beyond one node (because tra= ffic will never be high, you will never have much data or you never need= the extra availability) then you of course could use one single node wi= thout lots of RAM, but it _could_ also be that *SQL is a better option?= =C2=A0

If you are a startup with great plans= but no money and not much data/traffic (yet:), or just want to test Cas= sandra, then I still would suggest >1 nodes (3?), and instead reduce= RAM and/or CPU to get the right price. =C2=A0In that case you could try= to use =C2=A0Tomcat on one of the nodes. =C2=A0This way you can see and= test what Cassandra does really well. =C2=A0It is not really that Cassa= ndra demands more than other databases, but more that the problems it is= designed to solve demands more than some other problems. =C2=A0FYI, I u= se 2-3 nodes with 2gb RAM each for testing, and have never had any probl= ems with that. =C2=A0

.vegard,

----- Original Message -----
From: user@cassandra.apache.org

To:
"user" <user@cassandra.apache.org>
Cc:

Sent:<= /div>Wed, 18 Sep 2013 19:04:27 +0530
Subject:
Why don't you start off with a =E2=80=9Csingle & sm= all=E2=80=9D Cassandra server as you usually do it with MySQL?

=0A

=0AFor any website just starting out, the load initiall= y is minimal & grows with a slow pace initially. People usually star= t with their MySQL based sites with a single server(***that too a VPS no= t a dedicated server) running as both app server as well as DB server &a= mp; usually get too far with this setup & only as they feel the need= they separate the DB from the app server giving it a separate VPS serve= r. This is what a start up expects the things to be while planning about= resources procurement.

=0A

=0ABut so far what I have seen, it's something very di= fferent with Cassandra. People usually recommend starting out with atlea= st a 3 node cluster, (on dedicated servers) with lots & lots of RAM.= 4GB or 8GB RAM is what they suggest to start with. So is it that Cassan= dra requires more hardware resources in comparison to MySQL, for a websi= te to deliver similar performance, serve similar load/ traffic & sam= e amount of data. I understand about higher storage requirements of Cass= andra due to replication but what about other hardware resources ?

= =0A

=0ACan't= we start off with Cassandra based apps just like MySQL. Starting with 1= or 2 VPS & adding more whenever there's a need ?

=0A

=0AI don't want to compa= re apples with oranges. I just want to know how much more dangerous situ= ation I may be in when I start out with a single node VPS based cassandr= a installation Vs a single node VPS based MySQL installation. Difference= between these two situations. Are cassandra servers more prone to be un= available than MySQL servers ? What is bad if I put tomcat too along wit= h Cassandra as people use LAMP stack on single server.

=0A

=0A---------

=0A

=0A

=0A=

=0AThis q= uestion is also posted at StackOverflow here=C2=A0& has an open bounty worth= +50 rep.

=0A
=0A
--=_d491296a1c250e59f5b6d14ae1ac923a--