Mailing-List: contact user-help@kudu.apache.org; run by ezmlm
Precedence: bulk
Reply-To: user@kudu.apache.org
MIME-Version: 1.0
In-Reply-To: <CAO3rhrSxfwXKaKSEAgzf4zRuYts7PYP6q-1zx4Kyb1D5UPbipA@mail.gmail.com>
References: <CAO3rhrSxfwXKaKSEAgzf4zRuYts7PYP6q-1zx4Kyb1D5UPbipA@mail.gmail.com>
From: David Alves <davidralves@gmail.com>
Date: Thu, 27 Apr 2017 10:50:16 -0700
Message-ID: <CAAnRTpJXUdEMobXpZYjaQ7MP9phDM0ZzxOxs7=n-jaLcp+KLmA@mail.gmail.com>
Subject: Re: Kudu Table Design Question
To: user <user@kudu.apache.org>
Content-Type: multipart/alternative; boundary=94eb2c1921ea24c666054e2998fd
archived-at: Thu, 27 Apr 2017 17:50:41 -0000

--94eb2c1921ea24c666054e2998fd
Content-Type: text/plain; charset=UTF-8

The suggestion of 20-30 per tablet is more about the number of available
cores than the size of the data.
Tools like impala derive parallelism from the number of tablets thus having
that count adjusted (but not necessarily equal to) to the core count gives
you a good performance tradeoff.
Of course this is not a hard limit, the tablet server should be able
anything up to 100 reasonably well depending on your hardware.

HTH
-david


On Tue, Apr 25, 2017 at 9:19 PM, Jason Heo <jason.heo.sde@gmail.com> wrote:

> Hi.
>
> This email
> <http://mail-archives.apache.org/mod_mbox/kudu-user/201702.mbox/%3CCALo2W-WkicyuSiErfn2bNPyDAVd%3DjxA0hLij%2BA_tdtVBTYu-wQ%40mail.gmail.com%3E> (and
> many other resources) suggests that tserver should have small number of
> tablets.
>
> In the above mail, Dan says that:
>
> >> something more like 20 or 30 would be ideal depending on hardware..
> >> ...
> >> I would aim for tablet size on the order of 50GiB,
>
>
> I'm curious why 20~30 per tserver would be ideal. Does this mean storing
> 1TB~1.5TB per tserver is ideal? Could someone please explain this?
>
> The reason I ask is that I'm currently doing capacity planning.
>
> Regards,
>
> Jason
>

--94eb2c1921ea24c666054e2998fd
Content-Type: text/html; charset=UTF-8
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr">The suggestion of 20-30 per tablet is more about the numbe=
r of available cores than the size of the data.<div>Tools like impala deriv=
e parallelism from the number of tablets thus having that count adjusted (b=
ut not necessarily equal to) to the core count gives you a good performance=
 tradeoff.</div><div>Of course this is not a hard limit, the tablet server =
should be able anything up to 100 reasonably well depending on your hardwar=
e.</div><div><br></div><div>HTH</div><div>-david</div><div><br></div></div>=
<div class=3D"gmail_extra"><br><div class=3D"gmail_quote">On Tue, Apr 25, 2=
017 at 9:19 PM, Jason Heo <span dir=3D"ltr">&lt;<a href=3D"mailto:jason.heo=
.sde@gmail.com" target=3D"_blank">jason.heo.sde@gmail.com</a>&gt;</span> wr=
ote:<br><blockquote class=3D"gmail_quote" style=3D"margin:0 0 0 .8ex;border=
-left:1px #ccc solid;padding-left:1ex"><div dir=3D"ltr">Hi.<div><br></div><=
div>This <a href=3D"http://mail-archives.apache.org/mod_mbox/kudu-user/2017=
02.mbox/%3CCALo2W-WkicyuSiErfn2bNPyDAVd%3DjxA0hLij%2BA_tdtVBTYu-wQ%40mail.g=
mail.com%3E" target=3D"_blank">email</a>=C2=A0(and many other resources) su=
ggests that tserver should have small number of tablets.</div><div><br></di=
v><div>In the above mail, Dan says that:</div><blockquote>&gt;&gt; somethin=
g more like 20 or 30 would be ideal depending on hardware..<br>&gt;&gt; ...=
<br>&gt;&gt; I would aim for tablet size on the order of 50GiB,</blockquote=
><div><br></div><div>I&#39;m curious why 20~30 per tserver would be ideal. =
Does this mean storing 1TB~1.5TB per tserver is ideal? Could someone please=
 explain this?</div><div><br></div><div>The reason I ask is that I&#39;m cu=
rrently doing capacity planning.</div><div><br></div><div>Regards,</div><di=
v><br></div><div>Jason</div></div>
</blockquote></div><br></div>

--94eb2c1921ea24c666054e2998fd--