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 john@disqus.com designates
 209.85.210.177 as permitted sender)
MIME-Version: 1.0
Date: Sun, 28 Apr 2013 14:21:20 -0700
Message-ID: 
 <CA+qt5VM9-p-GUVS6mv=r5FGrdu5PL+kfdbuQX2Rphv527c-3Wg@mail.gmail.com>
Subject: cassandra-shuffle time to completion and required disk space
From: John Watson <john@disqus.com>
To: user@cassandra.apache.org
Content-Type: multipart/alternative; boundary=047d7b10d0b571fadf04db72598c

--047d7b10d0b571fadf04db72598c
Content-Type: text/plain; charset=UTF-8

The amount of time/space cassandra-shuffle requires when upgrading to using
vnodes should really be apparent in documentation (when some is made).

Only semi-noticeable remark about the exorbitant amount of time is a bullet
point in: http://wiki.apache.org/cassandra/VirtualNodes/Balance

"Shuffling will entail moving a lot of data around the cluster and so has
the potential to consume a lot of disk and network I/O, and to take a
considerable amount of time. For this to be an online operation, the
shuffle will need to operate on a lower priority basis to other streaming
operations, and should be expected to take days or weeks to complete."

We tried running shuffle on a QA version of our cluster and 2 things were
brought to light:
 - Even with no reads/writes it was going to take 20 days
 - Each machine needed enough free diskspace to potentially hold the entire
cluster's sstables on disk

Regards,

John

--047d7b10d0b571fadf04db72598c
Content-Type: text/html; charset=UTF-8
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr">The amount of time/space cassandra-shuffle requires when u=
pgrading to using vnodes should really be apparent in documentation (when s=
ome is made).<div><br></div><div style>Only semi-noticeable remark about th=
e exorbitant amount of time is a bullet point in:=C2=A0<a href=3D"http://wi=
ki.apache.org/cassandra/VirtualNodes/Balance">http://wiki.apache.org/cassan=
dra/VirtualNodes/Balance</a></div>
<div style><br></div><div style>&quot;Shuffling will entail moving a lot of=
 data around the cluster and so has the potential to consume a lot of disk =
and network I/O, and to take a considerable amount of time. For this to be =
an online operation, the shuffle will need to operate on a lower priority b=
asis to other streaming operations, and should be expected to take days or =
weeks to complete.&quot;</div>
<div style><br></div><div style>We tried running shuffle on a QA version of=
 our cluster and 2 things were brought to light:</div><div style>=C2=A0- Ev=
en with no reads/writes it was going to take 20 days</div><div style>=C2=A0=
- Each machine needed enough free diskspace to potentially hold the entire =
cluster&#39;s sstables on disk</div>
<div style><br></div><div style>Regards,</div><div style><br></div><div sty=
le>John</div></div>

--047d7b10d0b571fadf04db72598c--