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: domain of yuzhihong@gmail.com designates
 209.85.220.53 as permitted sender)
References: <1405288475243-4061296.post@n3.nabble.com>
Mime-Version: 1.0 (1.0)
In-Reply-To: <1405288475243-4061296.post@n3.nabble.com>
Content-Type: text/plain;
	charset=us-ascii
Content-Transfer-Encoding: quoted-printable
Message-Id: <0474165B-52FF-4CEA-A673-32D66CFE101E@gmail.com>
Cc: "user@hbase.apache.org" <user@hbase.apache.org>
From: Ted Yu <yuzhihong@gmail.com>
Subject: Re: Is there an advantage of placing related regions on the same
 region server/host?
Date: Sun, 13 Jul 2014 15:59:59 -0700
To: "user@hbase.apache.org" <user@hbase.apache.org>

By related regions, do you mean query would typically involve both regions ?=


Cheers

On Jul 13, 2014, at 2:54 PM, gomes <sankarmail@gmail.com> wrote:

> I have few tables, and I have pre partitioned the tables. I would like to
> achieve reduced latency. I am just thinking of placing related regions on
> the same region server/host. For example, table 1 contains 3 regions
> [a,b,c], and table 2 contains 3 regions[x,y,z]. (a,x) contains the data fo=
r
> some group of users, (b,y) contains the data for some group of users, and
> (c,z) and so on.=20
>=20
> If I have two region servers, instead of placing [a,b,c] on one server, an=
d
> [x,y,z] on other server, is there an advantage of placing related regions
> (a,x), (b,y), (c,z) among the region servers?
>=20
> Note: I also make sure the regions will not be divided further, or there
> will be manual intervention needed to divide the regions further.
>=20
>=20
>=20
> --
> View this message in context: http://apache-hbase.679495.n3.nabble.com/Is-=
there-an-advantage-of-placing-related-regions-on-the-same-region-server-host=
-tp4061296.html
> Sent from the HBase User mailing list archive at Nabble.com.