Mailing-List: contact user-help@cassandra.apache.org; run by ezmlm
Precedence: bulk
Reply-To: user@cassandra.apache.org
Received-SPF: pass (nike.apache.org: domain of jbellis@gmail.com designates
 74.125.82.44 as permitted sender)
DomainKey-Signature: a=rsa-sha1; c=nofws;
        d=gmail.com; s=gamma;
        h=mime-version:in-reply-to:references:from:date:message-id:subject:to
         :cc:content-type:content-transfer-encoding;
        b=kcZwuP/cuUKqZm2c4CExe5YlyvyWkPEUOPMeUMaMZofnhrgMtQ3nOllQoMx3rBP8hV
         i+cLD9/bPCrHg3qHf+wAkgFymkQsKzmAXETePKSgJ8HxUHN7SU6gh6m82AffgGKP8B7i
         +gAwvbFJ3l/gIuuWn0u+fRqilCthJvOOt1TbQ=
MIME-Version: 1.0
In-Reply-To: <BANLkTimizYk6iCs4P9dp-UnRC+QSbN7eAQ@mail.gmail.com>
References: <BANLkTimizYk6iCs4P9dp-UnRC+QSbN7eAQ@mail.gmail.com>
From: Jonathan Ellis <jbellis@gmail.com>
Date: Thu, 12 May 2011 18:14:30 -0500
Message-ID: <BANLkTinvOSNT8un_n+ocp+A34MnwdDNtdQ@mail.gmail.com>
Subject: Re: running TPC-C on cassandra clusters
To: user@cassandra.apache.org
Cc: Jingxin Feng <fengjingxin@gmail.com>
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable

I'd recommend trying the 0.7 svn branch (soon to be voted on as 0.7.6)

On Thu, May 12, 2011 at 3:09 PM, Xiaowei Wang <xiaowei609@gmail.com> wrote:
> Hi all,
>
> My partner and I currently using cassandra cluster to run TPC-C. We first
> use 2 ec2 nodes to load 20 warehouses. One(client node)=A0 has 8 cores,=
=A0 the
> other(worker node) has 4 cores. During the loading time, either the clien=
t
> node or the worker node will "down"(cannot be detected) randomly and then
> "up" again in a short time. If the two nodes both down, we failed in
> loading. If only one of them down, we can continue to load data.
>
> The problem is if we use multiple threads(we write multiprocess code), sa=
y 4
> clients threads, some of them might be stop at the point one of the nodes
> first down, and the dead threads will never come back.... This will not o=
nly
> enlarge our loading time, but also effect the amount of data we can load.
>
> So we need to figure out why the nodes continue to be up and down and fix
> this problem.
>
> Thanks for any help!
>
> Best,
> Xiaowei
>
>


--=20
Jonathan Ellis
Project Chair, Apache Cassandra
co-founder of DataStax, the source for professional Cassandra support
http://www.datastax.com