Mailing-List: contact user-help@cassandra.apache.org; run by ezmlm
Precedence: bulk
Reply-To: user@cassandra.apache.org
Received-SPF: pass (athena.apache.org: domain of stuhood@mailtrust.com
 designates 207.97.245.151 as permitted sender)
Date: Wed, 26 May 2010 12:01:32 -0500 (CDT)
Subject: RE: using more than 50% of disk space
From: "Stu Hood" <stu.hood@rackspace.com>
To: user@cassandra.apache.org
MIME-Version: 1.0
Content-Type: text/plain;charset=UTF-8
Content-Transfer-Encoding: quoted-printable
Importance: Normal
In-Reply-To: <AANLkTileksl4A6MotaRhGTy1SHMpqQWUg8rmOYirOduZ@mail.gmail.com>
References: <AANLkTileksl4A6MotaRhGTy1SHMpqQWUg8rmOYirOduZ@mail.gmail.com>
Message-ID: <1274893292.0321129@192.168.2.228>

See https://issues.apache.org/jira/browse/CASSANDRA-579 for some background=
 here: I was just about to start working on this one, but it won't make it =
in until 0.7.=0A=0A=0A-----Original Message-----=0AFrom: "Sean Bridges" <se=
an.bridges@gmail.com>=0ASent: Wednesday, May 26, 2010 11:50am=0ATo: user@ca=
ssandra.apache.org=0ASubject: using more than 50% of disk space=0A=0AWe're =
investigating Cassandra, and we are looking for a way to get Cassandra=0Aus=
e more than 50% of it's data disks.  Is this possible?=0A=0AFor major compa=
ctions, it looks like we can use more than 50% of the disk if=0Awe use mult=
iple similarly sized column families.  If we had 10 column=0Afamilies of th=
e same size, we could use 90% of the disk, since a major=0Acompaction would=
 only need as much free space as the largest column family=0A(in reality we=
 would use less).  Is that right?=0A=0AFor bootstrapping new nodes, it look=
s like adding a new node will require=0Athat an existing node does anti-com=
paction.  This anti-compaction could take=0Anearly 50% of the disk.  Is the=
re a way around this?=0A=0AIs there anything else that would prevent us fro=
m using more than 50% of the=0Adata disk.=0A=0AThanks,=0A=0ASean=0A