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-Mailer:Message-ID:Date:From:Reply-To:Subject:To:MIME-Version:Content-Type;
  b=cJNaN1I3Mb3iQAGxmiH4wq92BhuiIftXkvF4Ox735ebLTI972a1/ZrKuqllKNtG3QEJMCOqV4T4CYtJgkbaeiM4dzFoyV0UAyXTcd6n9qKr56G6XWzn3Iyo9XJHyWvV/yHZuY281TQOekObzxSk3fNL4LgvGLXJUNN4mpQyguek=;
Message-ID: <1337927705.87668.YahooMailNeo@web132104.mail.ird.yahoo.com>
Date: Fri, 25 May 2012 07:35:05 +0100 (BST)
From: Florin P <florinpico@yahoo.com>
Reply-To: Florin P <florinpico@yahoo.com>
Subject: A question about HBase MapReduce
To: "user@hbase.apache.org" <user@hbase.apache.org>
MIME-Version: 1.0
Content-Type: multipart/alternative;
 boundary="-1910628564-2145564354-1337927705=:87668"

---1910628564-2145564354-1337927705=:87668
Content-Type: text/plain; charset=iso-8859-1
Content-Transfer-Encoding: quoted-printable

Hello!=0A=0AI've read Lars George's blog http://www.larsgeorge.com/2009/05/=
hbase-mapreduce-101-part-i.html where at the end of the article, he mention=
ed "In the next post I will show you how to import data from a raw data=0Af=
ile into a HBase table and how you eventually process the data in the=0AHBa=
se table. We will address questions like how many mappers and/or=0Areducers=
 are needed and how can I improve import and processing=0Aperformance.". I =
looked in the blog up for these questions, but it seems that there is no ar=
ticle related. Do you knoe if he you touched these subjects into a differen=
t post or book? Particular I am interested=A0 =0A=0A1. how you can set up t=
he number of mappers?=0A2. number of mappers can be set up per region serve=
r? If yes how?=0A3. How the big number of set up mappers can affect the dat=
a locality?=0A4. is this algorithm for computing the number of mappers (htt=
ps://issues.apache.org/jira/browse/HBASE-1172) still available=0A"Currently=
,=0Athe number of mappers specified when using TableInputFormat is strictly=
=0Afollowed if less than total regions on the input table. If greater, the=
=0Anumber of regions is used.=0AThis will modify the splitting algorithm to=
 do the following:=0A=09* Specify 0 mappers when you want # mappers =3D # r=
egions=0A=09* If you specify fewer mappers than regions, will use exactly t=
he number you specify based on the current algorithm=0A=09* If=0Ayou specif=
y more mappers than regions, will divide regions up by=0Adetermining [start=
,X) [X,end). The number of mappers will always be a=0Amultiple of number of=
 regions. This is so we do not have scanners=0Aspanning multiple regions.=
=0AThere is an additional issue in that the default number of mappers=0Ain =
JobConf is set to 1. That means if a user does not explicitly set=0Anumber =
of map tasks, a single mapper will be used. "=0A=0AI'll look forward for yo=
u answers. Thank you.=0A=0AKind regards,=A0Florin
---1910628564-2145564354-1337927705=:87668--