Mailing-List: contact cassandra-user-help@incubator.apache.org; run by ezmlm
Precedence: bulk
Reply-To: cassandra-user@incubator.apache.org
Received-SPF: pass (nike.apache.org: domain of eprosenx@gmail.com designates
 209.85.218.213 as permitted sender)
DomainKey-Signature: a=rsa-sha1; c=nofws;
        d=gmail.com; s=gamma;
        h=mime-version:date:message-id:subject:from:to:content-type;
        b=K8x59K157yEeXu2dRHOvlU7ijN2VuaTJJzGJo5SMfzKFpYQUmqElctNXGLmdFZu3/x
         krfKFIqzt8FCT2osuZU4JWn1aJnH457bEIRt/Iwmef7hR/YB7qADCTQL2wDmQ4pvNSkA
         Xwa0z4yK2OU1Qfq7C5YPaiC4OyUDvmTycC6sw=
MIME-Version: 1.0
Date: Wed, 10 Mar 2010 00:00:23 -0800
Message-ID: <a409e0b51003100000j1083cb9v35cf43cb323abcb@mail.gmail.com>
Subject: Effective allocation of multiple disks
From: Eric Rosenberry <eprosenx@gmail.com>
To: cassandra-user@incubator.apache.org
Content-Type: multipart/alternative; boundary=000325559052c2d8b404816db081

--000325559052c2d8b404816db081
Content-Type: text/plain; charset=ISO-8859-1

Based on the documentation, it is clear that with Cassandra you want to have
one disk for commitlog, and one disk for data.

My question is: If you think your workload is going to require more io
performance to the data disks than a single disk can handle, how would you
recommend effectively utilizing additional disks?

It would seem a number of vendors sell 1U boxes with four 3.5 inch disks.
 If we use one for commitlog, is there a way to have Cassandra itself
equally split data across the three remaining disks?  Or is this something
that needs to be handled by the hardware level, or operating system/file
system level?

Options include a hardware RAID controller in a RAID 0 stripe (this is more
$$$ and for what gain?), or utilizing a volume manager like LVM.

Along those same lines, if you do implement some type of striping, what RAID
stripe size is recommended?  (I think Todd Burruss asked this earlier but I
did not see a response)

Thanks for any input!

-Eric

--000325559052c2d8b404816db081
Content-Type: text/html; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable

Based on the documentation, it is clear that with Cassandra you want to hav=
e one disk for commitlog, and one disk for data.<div><br></div><div>My ques=
tion is: If you think your workload is going to require more io performance=
 to the data disks than a single disk can handle, how would you recommend e=
ffectively utilizing additional disks?</div>
<div><br></div><div>It would seem a number of vendors sell 1U boxes with fo=
ur 3.5 inch disks. =A0If we use one for commitlog, is there a way to have C=
assandra itself equally split data across the three remaining disks? =A0Or =
is this something that needs to be handled by the hardware level, or operat=
ing system/file system level?</div>
<div><br></div><div>Options include a hardware RAID controller in a RAID 0 =
stripe (this is more $$$ and for what gain?), or utilizing a volume manager=
 like LVM.</div><div><br></div><div>Along those same lines, if you do imple=
ment some type of striping, what RAID stripe size is recommended? =A0(I thi=
nk=A0Todd Burruss asked this earlier but I did not see a response)</div>
<div><br></div><div>Thanks for any input!<br><div><br></div><div>-Eric</div=
></div>

--000325559052c2d8b404816db081--