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 graham@vast.com designates
 209.85.219.41 as permitted sender)
Subject: 
 =?utf-8?Q?Re:_Does_SELECT_=E2=80=A6_IN_=28=29_use_parallel_dispatch=3F?=
References: 
 <CAAZU44mjXk+e_1C0Zet-3_Yc5CYznDykrxHhBvRM6dWAd5A-Sw@mail.gmail.com>
 <CABNXB2A+ctX_R3sh_A8zEzbKOBrzkXwwWUnh_F6YhxjnAo0cpg@mail.gmail.com>
From: Graham Sanderson <graham@vast.com>
Content-Type: multipart/alternative;
	boundary=Apple-Mail-2FD07EA8-C417-48BB-A556-D071DF230D4D
In-Reply-To: 
 <CABNXB2A+ctX_R3sh_A8zEzbKOBrzkXwwWUnh_F6YhxjnAo0cpg@mail.gmail.com>
Message-Id: <005D6653-0F50-496B-9022-E63947E114F1@vast.com>
Date: Fri, 25 Jul 2014 13:40:06 -0500
To: "user@cassandra.apache.org" <user@cassandra.apache.org>
Content-Transfer-Encoding: 7bit
Mime-Version: 1.0 (1.0)


--Apple-Mail-2FD07EA8-C417-48BB-A556-D071DF230D4D
Content-Type: text/plain;
	charset=utf-8
Content-Transfer-Encoding: quoted-printable

Of course the driver in question is allowed to be smarter and can do so if u=
se use a ? parameter for a list or even individual elements

I'm not sure which if any drivers currently do this but we plan to combine t=
his with token aware routing in our scala driver in the future=20

Sent from my iPhone

> On Jul 25, 2014, at 1:14 PM, DuyHai Doan <doanduyhai@gmail.com> wrote:
>=20
> Nope. Select ... IN() sends one request to a coordinator. This coordinator=
 dispatch the request to 50 nodes as in your example and waits for 50 respon=
ses before sending back the final result. As you can guess this approach is n=
ot optimal since the global request latency is bound to the slowest latency a=
mong 50 nodes.
>=20
>  On the other hand if you use async feature from the native protocol, you c=
lient will issue 50 requests in parallel and the answers arrive as soon as t=
hey are fetched from different nodes.
>=20
>  Clearly the only advantage of using IN() clause is ease of query. I would=
 advise to use IN() only when you have a "few" values, not 50.
>=20
>=20
>> On Fri, Jul 25, 2014 at 8:08 PM, Kevin Burton <burton@spinn3r.com> wrote:=

>> Say I have about 50 primary keys I need to fetch.
>>=20
>> I'd like to use parallel dispatch.  So that if I have 50 hosts, and each h=
as record, I can read from all 50 at once.
>>=20
>> I assume cassandra does the right thing here ?  I believe it does=E2=80=A6=
 at least from reading the docs but it's still a bit unclear.
>>=20
>> Kevin
>>=20
>> --=20
>> Founder/CEO Spinn3r.com
>> Location: San Francisco, CA
>> blog: http://burtonator.wordpress.com
>> =E2=80=A6 or check out my Google+ profile
>>=20
>=20

--Apple-Mail-2FD07EA8-C417-48BB-A556-D071DF230D4D
Content-Type: text/html;
	charset=utf-8
Content-Transfer-Encoding: quoted-printable

<html><head><meta http-equiv=3D"content-type" content=3D"text/html; charset=3D=
utf-8"></head><body dir=3D"auto"><div>Of course the driver in question is al=
lowed to be smarter and can do so if use use a ? parameter for a list or eve=
n individual elements</div><div><br></div><div>I'm not sure which if any dri=
vers currently do this but we plan to combine this with token aware routing i=
n our scala driver in the future&nbsp;<br><br>Sent from my iPhone</div><div>=
<br>On Jul 25, 2014, at 1:14 PM, DuyHai Doan &lt;<a href=3D"mailto:doanduyha=
i@gmail.com">doanduyhai@gmail.com</a>&gt; wrote:<br><br></div><blockquote ty=
pe=3D"cite"><div><div dir=3D"ltr">Nope. Select ... IN() sends one request to=
 a coordinator. This coordinator dispatch the request to 50 nodes as in your=
 example and waits for 50 responses before sending back the final result. As=
 you can guess this approach is not optimal since the global request latency=
 is bound to the slowest latency among 50 nodes.<div>
<br></div><div>&nbsp;On the other hand if you use async feature from the nat=
ive protocol, you client will issue 50 requests in parallel and the answers a=
rrive as soon as they are fetched from different nodes.</div><div><br></div>=

<div>&nbsp;Clearly the only advantage of using IN() clause is ease of query.=
 I would advise to use IN() only when you have a "few" values, not 50.</div>=
</div><div class=3D"gmail_extra"><br><br><div class=3D"gmail_quote">
On Fri, Jul 25, 2014 at 8:08 PM, Kevin Burton <span dir=3D"ltr">&lt;<a href=3D=
"mailto:burton@spinn3r.com" target=3D"_blank">burton@spinn3r.com</a>&gt;</sp=
an> wrote:<br><blockquote class=3D"gmail_quote" style=3D"margin:0 0 0 .8ex;b=
order-left:1px #ccc solid;padding-left:1ex">
<div dir=3D"ltr">Say I have about 50 primary keys I need to fetch.<div><br><=
/div><div>I'd like to use parallel dispatch. &nbsp;So that if I have 50 host=
s, and each has record, I can read from all 50 at once.</div><div><br></div>=


<div>I assume cassandra does the right thing here ? &nbsp;I believe it does=E2=
=80=A6 at least from reading the docs but it's still a bit unclear.</div><sp=
an class=3D"HOEnZb"><font color=3D"#888888"><div><br></div><div>Kevin<br cle=
ar=3D"all">
<div><br></div>-- <br><div dir=3D"ltr">
<div>
<div><p style=3D"margin-top:0px;margin-right:0px;margin-bottom:12pt;margin-l=
eft:0px"></p><div>Founder/CEO&nbsp;<a href=3D"http://Spinn3r.com" target=3D"=
_blank">Spinn3r.com</a><br></div><div>Location:&nbsp;<b>San Francisco, CA</b=
><br></div>


<div><font color=3D"#2c2c2c" face=3D"Helvetica, Arial, sans-serif"><span sty=
le=3D"line-height:19px">blog:<b>&nbsp;</b></span></font><a href=3D"http://bu=
rtonator.wordpress.com" target=3D"_blank">http://burtonator.wordpress.com</a=
></div>

<div>
=E2=80=A6 or check out my <a href=3D"https://plus.google.com/102718274791889=
610666/posts" target=3D"_blank">Google+ profile</a></div><div><a href=3D"htt=
p://spinn3r.com" target=3D"_blank"><img src=3D"http://spinn3r.com/images/spi=
nn3r.jpg"></a></div>


<p></p></div></div></div>
</div></font></span></div>
</blockquote></div><br></div>
</div></blockquote></body></html>=

--Apple-Mail-2FD07EA8-C417-48BB-A556-D071DF230D4D--