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 kohlisankalp@gmail.com
 designates 209.85.216.47 as permitted sender)
MIME-Version: 1.0
In-Reply-To: <1372358454.85687.YahooMailNeo@web142804.mail.bf1.yahoo.com>
References: <1372358454.85687.YahooMailNeo@web142804.mail.bf1.yahoo.com>
From: sankalp kohli <kohlisankalp@gmail.com>
Date: Thu, 27 Jun 2013 15:44:39 -0400
Message-ID: 
 <CAEPxca19Omua2bTt2_So=64U1QXaXL_wcGL7yvXwKD1rnee6qQ@mail.gmail.com>
Subject: Re: High-read latency for non-existing rows with LCS and 1.2.5
To: user@cassandra.apache.org, Bao Le <lvqb@yahoo.com>
Content-Type: multipart/alternative; boundary=90e6ba10b0018d7e5c04e02800ad

--90e6ba10b0018d7e5c04e02800ad
Content-Type: text/plain; charset=ISO-8859-1

Try doing request tracing.
http://www.datastax.com/dev/blog/tracing-in-cassandra-1-2


On Thu, Jun 27, 2013 at 2:40 PM, Bao Le <lvqb@yahoo.com> wrote:

> Hi,
>
>     We are using Leveled Compaction with Cassandra 1.2.5. Our sstable size
> is 100M. On each node,
> we have anywhere from 700+ to 800+ sstables (for all levels). The
> bloom_filter_fp_chance is set at 0.000744.
>
>   For read requests that ask for existing row records, the latency is
> great, mostly under 20 milliseconds with key cache and row cache set. For
> read requests that ask for non-existing row records (not because of
> deletes, but rather, have never been in the system to start with), the
> latency is running into hundreds of milliseconds and sometimes seconds.
>
>   Just wonder if anyone has come across this before and has some pointers
> on how to reduce the latency in this case.
>
> Thanks
> Bao
>
>
>
>

--90e6ba10b0018d7e5c04e02800ad
Content-Type: text/html; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr">Try doing request tracing.=A0<div><a href=3D"http://www.da=
tastax.com/dev/blog/tracing-in-cassandra-1-2">http://www.datastax.com/dev/b=
log/tracing-in-cassandra-1-2</a><br></div></div><div class=3D"gmail_extra">=
<br><br>

<div class=3D"gmail_quote">On Thu, Jun 27, 2013 at 2:40 PM, Bao Le <span di=
r=3D"ltr">&lt;<a href=3D"mailto:lvqb@yahoo.com" target=3D"_blank">lvqb@yaho=
o.com</a>&gt;</span> wrote:<br><blockquote class=3D"gmail_quote" style=3D"m=
argin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">

<div><div style=3D"font-size:12pt;font-family:times new roman,new york,time=
s,serif">Hi,<br><br>=A0=A0=A0 We are using Leveled Compaction with Cassandr=
a 1.2.5. Our sstable size is 100M. On each node,<br>we have anywhere from 7=
00+ to 800+ sstables (for all levels). The bloom_filter_fp_chance is set at=
 0.000744.<br>

<br>=A0 For read requests that ask for existing row records, the latency is=
 great, mostly under 20 milliseconds with key cache and row cache set. For =
read requests that ask for non-existing row records (not because of deletes=
, but rather, have never been in the system to start with), the latency is =
running into hundreds of milliseconds and sometimes seconds.<br>

<br>=A0 Just wonder if anyone has come across this before and has some poin=
ters on how to reduce the latency in this case.<br><br>Thanks<span class=3D=
"HOEnZb"><font color=3D"#888888"><br>Bao<br><br><br><div><br></div></font><=
/span></div>

</div></blockquote></div><br></div>

--90e6ba10b0018d7e5c04e02800ad--