Mailing-List: contact user-help@hbase.apache.org; run by ezmlm
Precedence: bulk
Reply-To: user@hbase.apache.org
Received-SPF: pass (nike.apache.org: local policy)
DomainKey-Signature: a=rsa-sha1; q=dns; c=nofws;
  s=s1024; d=yahoo.com;
  h=X-YMail-OSG:Received:X-Rocket-MIMEInfo:X-RocketYMMF:X-Mailer:References:Message-ID:Date:From:Reply-To:Subject:To:In-Reply-To:MIME-Version:Content-Type;
  b=jJYveCNk/uTlo87wevH89s+ugObozXMh+zaT5lHwSu8cc58ElzrkNOmlzGOOCkqr6+W43RemmljfN1VGnEnwB4ByHUp6Om6HD1GBXlqdhLb9Mduo7irwcnwY2yxGn85F1WE/fKPYvwv1/H2M6GrYJ97tbmvT6XZCHlDiF6Lkl6U=;
References: 
 <CAG2rwuOCcoCk5NSVmzEecmCqg0n6eNEV9+rbS9KAyD6TV1Fa-g@mail.gmail.com>
Message-ID: <1358881380.57085.YahooMailNeo@web140602.mail.bf1.yahoo.com>
Date: Tue, 22 Jan 2013 11:03:00 -0800 (PST)
From: lars hofhansl <larsh@apache.org>
Reply-To: lars hofhansl <larsh@apache.org>
Subject: Re: write throughput in cassandra, understanding hbase
To: "user@hbase.apache.org" <user@hbase.apache.org>
In-Reply-To: 
 <CAG2rwuOCcoCk5NSVmzEecmCqg0n6eNEV9+rbS9KAyD6TV1Fa-g@mail.gmail.com>
MIME-Version: 1.0
Content-Type: multipart/alternative;
 boundary="-118416272-892707874-1358881380=:57085"

---118416272-892707874-1358881380=:57085
Content-Type: text/plain; charset=iso-8859-1
Content-Transfer-Encoding: quoted-printable

Where do you see that HBase is doing only 2-3k writes/s?=0AHow was the data=
 distributed? Was the table split?=0ACassandra uses a random partitioner by=
 default, which will nicely distribute the data over the cluster but won't =
allow to perform range scans over your data.=0AHBase always partitions by k=
ey ranges, so that the keys can the range scanned. If that is not done corr=
ectly and you create monotonically increasing keys, you'll hotspot a single=
 region server.=0A=0AEven then, you can do more than this on single RegionS=
erver.=0A=0AAlso note that many of the benchmarks have agendas and cherry p=
ick the results.=0AThey probably "forgot" to disabled Nagle's and to distri=
bute the table correctly.=0A=0A-- Lars=0A=0A=0A=0A_________________________=
_______=0A From: S Ahmed <sahmed1020@gmail.com>=0ATo: user@hbase.apache.org=
 =0ASent: Tuesday, January 22, 2013 10:38 AM=0ASubject: RE: write throughpu=
t in cassandra, understanding hbase=0A =0AI've read articles online where I=
 see cassandra doing like 20K writers per=0Asecond, and hbase around 2-3K.=
=0A=0AI understand both systems have their strenghts, but I am curious as t=
o what=0Ais holding hbase from reaching similiar results?=0A=0AIs it HDFS t=
hat is the issue?=A0 Or hbase does certain things (to its=0Aadvantage) that=
 slows the write path down?
---118416272-892707874-1358881380=:57085--