Mailing-List: contact cassandra-user-help@incubator.apache.org; run by ezmlm
Precedence: bulk
Reply-To: cassandra-user@incubator.apache.org
Received-SPF: pass (athena.apache.org: domain of michael.greene@gmail.com
 designates 209.85.216.204 as permitted sender)
DomainKey-Signature: a=rsa-sha1; c=nofws;
        d=gmail.com; s=gamma;
        h=mime-version:in-reply-to:references:date:message-id:subject:from:to
         :content-type;
        b=OQtER3LIrZhH4q5UB+pameQxxKI4B+Bgv0vuwZB2P8GKJyv89jPALu60XvLodBeDHE
         yHrfcTkqKHoIvK7IALwFqBtN+extXeAyVJRUAQz+F9HgUwmXpBD8V9X1K2snFS3/kZ1/
         q9tZTUyLnq52LgSqI8aSgKbJH8w7Bf8Uf0Asw=
MIME-Version: 1.0
In-Reply-To: <87ws3n120g.wl%catap@catap.ru>
References: <87ws3n120g.wl%catap@catap.ru>
Date: Fri, 25 Sep 2009 10:34:06 -0500
Message-ID: <f4d6a21a0909250834m7626985bp1e99281d7e4f9c98@mail.gmail.com>
Subject: Re: perfomance issue
From: Michael Greene <michael.greene@gmail.com>
To: cassandra-user@incubator.apache.org
Content-Type: multipart/alternative; boundary=000e0cd1723aafc10e047468ad51

--000e0cd1723aafc10e047468ad51
Content-Type: text/plain; charset=ISO-8859-1

Thanks for the results.  Perhaps you could shed further light:Is this a
single node system?
Is the log level changed from DEBUG to INFO?
Are the commit log and data directories on the same drive?
Are the sets/gets being processed interleaved in parallel, or one then the
other?

Note that writes are ~6x and ~5x slower with those values, and gets are ~50x
and ~20x slower

Michael

On Fri, Sep 25, 2009 at 10:23 AM, Kirill A. Korinskiy <
catap+cassandra@catap.ru <catap%2Bcassandra@catap.ru>> wrote:

> Hello.
>
> I'm write a small erlang benchmark: cassandra vs simple fs storage and
> I have a results:
>
> fs_storage:
> 10: 1732/228633/2873/1128 992/233261/2870/1026
> 25: 4531/35290/6084/632 1786/35292/5969/312
> 50: 4707/311825/13694/4644 1366/382913/14470/5823
>
> cassandra:
> 10: 69014/424051/145871/28066 9553/427823/141925/29190
> 25: 165500/536108/339736/77173 14846/710370/341819/78694
> 50: 325790/1050753/667931/157962 9865/1033461/676317/160606
>
> legends:
> parallel request: get set min/max/avg/mad
>
> So. Cassandra slower ~50 times.
>
> Ok I change cassandra setting:
>
>  ColumnFamily from UTF8Type to BytesType
>  MemtableObjectCountInMillions from 0.1 to 1
>  ConcurrentReads from 8 to 64
>  ConcurrentWrites from 32 to 64
>
> And I have:
>
> fs_storage:
> 10: 2164/245375/2515/474 651/247790/2746/800
> 25: 5521/19959/5838/196 1836/20138/5943/186
> 50: 6369/32837/11476/944 1396/32818/11449/876
>
> cassandra:
> 10: 39931/279903/48463/4812 21378/278782/51477/4526
> 25: 101911/407343/119135/9158 8794/410423/123768/9076
> 50: 209202/429959/238179/12180 7803/442580/244402/13443
>
> Cassandra slower ~20 times.
>
> I's normal? Can I do something wrong? Maybe you have any ideas?
>
> --
> wbr, Kirill
>

--000e0cd1723aafc10e047468ad51
Content-Type: text/html; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable

Thanks for the results. =A0Perhaps you could shed further light:<div>Is thi=
s a single node system?</div><div>Is the log level changed from DEBUG to IN=
FO?</div><div>Are the commit log and data directories on the same drive?</d=
iv>
<div>Are the sets/gets being processed interleaved in parallel, or one then=
 the other?</div><div><br></div><div>Note that writes are ~6x and ~5x slowe=
r with those values, and gets are ~50x and ~20x slower</div><div><br></div>
<div>Michael<br><br><div class=3D"gmail_quote">On Fri, Sep 25, 2009 at 10:2=
3 AM, Kirill A. Korinskiy <span dir=3D"ltr">&lt;<a href=3D"mailto:catap%2Bc=
assandra@catap.ru">catap+cassandra@catap.ru</a>&gt;</span> wrote:<br><block=
quote class=3D"gmail_quote" style=3D"margin:0 0 0 .8ex;border-left:1px #ccc=
 solid;padding-left:1ex;">
Hello.<br>
<br>
I&#39;m write a small erlang benchmark: cassandra vs simple fs storage and<=
br>
I have a results:<br>
<br>
fs_storage:<br>
10: 1732/228633/2873/1128 992/233261/2870/1026<br>
25: 4531/35290/6084/632 1786/35292/5969/312<br>
50: 4707/311825/13694/4644 1366/382913/14470/5823<br>
<br>
cassandra:<br>
10: 69014/424051/145871/28066 9553/427823/141925/29190<br>
25: 165500/536108/339736/77173 14846/710370/341819/78694<br>
50: 325790/1050753/667931/157962 9865/1033461/676317/160606<br>
<br>
legends:<br>
parallel request: get set min/max/avg/mad<br>
<br>
So. Cassandra slower ~50 times.<br>
<br>
Ok I change cassandra setting:<br>
<br>
=A0ColumnFamily from UTF8Type to BytesType<br>
=A0MemtableObjectCountInMillions from 0.1 to 1<br>
=A0ConcurrentReads from 8 to 64<br>
=A0ConcurrentWrites from 32 to 64<br>
<br>
And I have:<br>
<br>
fs_storage:<br>
10: 2164/245375/2515/474 651/247790/2746/800<br>
25: 5521/19959/5838/196 1836/20138/5943/186<br>
50: 6369/32837/11476/944 1396/32818/11449/876<br>
<br>
cassandra:<br>
10: 39931/279903/48463/4812 21378/278782/51477/4526<br>
25: 101911/407343/119135/9158 8794/410423/123768/9076<br>
50: 209202/429959/238179/12180 7803/442580/244402/13443<br>
<br>
Cassandra slower ~20 times.<br>
<br>
I&#39;s normal? Can I do something wrong? Maybe you have any ideas?<br>
<font color=3D"#888888"><br>
--<br>
wbr, Kirill<br>
</font></blockquote></div><br></div>

--000e0cd1723aafc10e047468ad51--