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 sylvain@datastax.com
 designates 209.85.214.172 as permitted sender)
MIME-Version: 1.0
In-Reply-To: <add86a36-b255-4ba6-95e2-1198a7257d5a@journal.report.generator>
References: <add86a36-b255-4ba6-95e2-1198a7257d5a@journal.report.generator>
Date: Fri, 30 Mar 2012 15:39:18 +0200
Message-ID: 
 <CAKkz8Q38VB=Zi48Z-Wq8=ef4PTD1bfZtNzRS06WbTfgrkcLE_g@mail.gmail.com>
Subject: Re: composite query performance depends on component ordering
From: Sylvain Lebresne <sylvain@datastax.com>
To: user@cassandra.apache.org
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable

When you do a query, there's a lot of comparison happening between
what's queries and the column names. But the composite comparator is
lazy in that when it compares two names, if the first component are
not equal, it doesn't have to compare the second one. So What's likely
happening is that in the first case you do 1 millions comparison of 8
bytes, in the latter you do 1 millions comparison of 100 bytes.

--
Sylvain

On Fri, Mar 30, 2012 at 2:30 PM, Alexandru Sicoe <adsicoe@gmail.com> wrote:
> Sender: adsicoe@gmail.com
> Subject: composite query performance depends on component ordering
> Message-Id: <CAADnm_f7xeznNMOp-fwgEy-f9uTjEPYNXdch3FXHqK6ncAggSQ@mail.gma=
il.com>
> Recipient: Adam.Nicholls@hl.co.uk
>
>
> ______________________________________________________________________
> This email has been scanned by the Symantec Email Security.cloud service.
> For more information please visit http://www.symanteccloud.com
> ______________________________________________________________________
>
> ---------- Forwarded message ----------
> From:=A0Alexandru Sicoe <adsicoe@gmail.com>
> To:=A0<user@cassandra.apache.org>
> Cc:
> Date:=A0Fri, 30 Mar 2012 14:29:47 +0200
> Subject:=A0composite query performance depends on component ordering
> Hi guys,
> =A0I am consistently seeing a 20% improvement in query retrieval times if=
 I use the composite comparator "Timestamp:ID" instead of "ID:Timestamp" wh=
ere Timestamp=3DLong and ID=3D~100 character strings. I am retrieving all c=
olumns=A0 (~1 million) from a single row. Why is this happening?
>
> Cheers,
> Alex
>