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
 74.125.82.172 as permitted sender)
MIME-Version: 1.0
In-Reply-To: <4E178AFD.7060604@bnl.gov>
References: <4E178AFD.7060604@bnl.gov>
From: Jonathan Ellis <jbellis@gmail.com>
Date: Fri, 8 Jul 2011 20:51:09 -0500
Message-ID: 
 <CALdd-zgzM2A-_1aLyfLRAn_8zjRVQkkxt=q4Cf4kuwLG3zoXCw@mail.gmail.com>
Subject: Re: Performance deterioration while building secondary index
To: user@cassandra.apache.org
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable

My guess: index build isn't respecting the background i/o throttle.

On Fri, Jul 8, 2011 at 5:55 PM, Maxim Potekhin <potekhin@bnl.gov> wrote:
> I have roughly 150 million rows in my database, which will grow as I
> continue testing. I'm building an index on a particular column, via
> cassandra-cli, something of the sort:
> update column family jobs with column_metadata =3D [{column_name : 'DATE'=
,
> =A0 validation_class : AsciiType, index_type : 0, index_name : 'date'}]
>
> At this point, the cluster just becomes unresponsive -- just doing "list"=
 on
> a CF takes a while. Random query test I used to run rather quickly, becom=
es
> terribly slow (hasn't returned since I started typing this).
>
> Is that normal? I can' imagine this happening in a production situation,
> when I decided to add an index for some valid reasons. Really scratching =
my
> head now. TIA.
>
> The version is 0.8.1
>
> Thanks!
>


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