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 arodrime@gmail.com designates
 209.85.219.50 as permitted sender)
MIME-Version: 1.0
In-Reply-To: <F3690910-E379-4BEB-BC76-29143D3832C8@ecyrd.com>
References: 
 <CA+VSrLrL_khhau28RKpL6TRAgBXpHDvhCfrfV6UhmZGQk5KK4w@mail.gmail.com>
 <F3690910-E379-4BEB-BC76-29143D3832C8@ecyrd.com>
From: Alain RODRIGUEZ <arodrime@gmail.com>
Date: Tue, 5 Mar 2013 17:50:06 +0100
Message-ID: 
 <CA+VSrLrbyS7rhX_6b=Ai2B+Fp-HHEqt4A3bfN1car=UmwUtb7A@mail.gmail.com>
Subject: Re: LCS and counters
To: user@cassandra.apache.org
Content-Type: multipart/alternative; boundary=e89a8fb1ee6e3ba78904d730459a

--e89a8fb1ee6e3ba78904d730459a
Content-Type: text/plain; charset=ISO-8859-1

Well no one says my assertion is false, so it is probably true.

Going further, what would be the steps to migrate from STC to LCS ? Is
there any precautions to take doing it using C*1.1.6 (like removing commit
logs since drain is broken) ?

Any insight or link on this procedure would be appreciated.


2013/2/25 Janne Jalkanen <Janne.Jalkanen@ecyrd.com>

>
> At least for our use case (reading slices from varyingly sized rows from
> 10-100k composite columns with counters and hundreds of writes/second) LCS
> has a nice ~75% lower read latency than Size Tiered. And compactions don't
> stop the world anymore.  Repairs do easily trigger a few hundred
> compactions though, but it's not that bad.
>
> /Janne
>
> On Feb 25, 2013, at 17:10 , Alain RODRIGUEZ <arodrime@gmail.com> wrote:
>
> > Hi
> >
> > I am just wondering... Wouldn't it always be worth it to use LCS on
> counter CF since LCS is optimized for reads and that writing a counter
> always require a read ?
> >
> > Alain
>
>

--e89a8fb1ee6e3ba78904d730459a
Content-Type: text/html; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr">Well no one says my assertion is false, so it is probably =
true.<div><br></div><div>Going further, what would be the steps to migrate =
from STC to LCS ? Is there any precautions to take doing it using C*1.1.6 (=
like removing commit logs since drain is broken) ?=A0</div>

<div><br></div><div>Any insight or link on this procedure would be apprecia=
ted.<br><div class=3D"gmail_extra"><br><br><div class=3D"gmail_quote">2013/=
2/25 Janne Jalkanen <span dir=3D"ltr">&lt;<a href=3D"mailto:Janne.Jalkanen@=
ecyrd.com" target=3D"_blank">Janne.Jalkanen@ecyrd.com</a>&gt;</span><br>

<blockquote class=3D"gmail_quote" style=3D"margin:0px 0px 0px 0.8ex;border-=
left-width:1px;border-left-color:rgb(204,204,204);border-left-style:solid;p=
adding-left:1ex"><br>
At least for our use case (reading slices from varyingly sized rows from 10=
-100k composite columns with counters and hundreds of writes/second) LCS ha=
s a nice ~75% lower read latency than Size Tiered. And compactions don&#39;=
t stop the world anymore. =A0Repairs do easily trigger a few hundred compac=
tions though, but it&#39;s not that bad.<br>


<span class=3D""><font color=3D"#888888"><br>
/Janne<br>
</font></span><div class=3D""><div class=3D"h5"><br>
On Feb 25, 2013, at 17:10 , Alain RODRIGUEZ &lt;<a href=3D"mailto:arodrime@=
gmail.com">arodrime@gmail.com</a>&gt; wrote:<br>
<br>
&gt; Hi<br>
&gt;<br>
&gt; I am just wondering... Wouldn&#39;t it always be worth it to use LCS o=
n counter CF since LCS is optimized for reads and that writing a counter al=
ways require a read ?<br>
&gt;<br>
&gt; Alain<br>
<br>
</div></div></blockquote></div><br></div></div></div>

--e89a8fb1ee6e3ba78904d730459a--