From: Scott Carey <scott@richrelevance.com>
To: Merlin Moncure <mmoncure@gmail.com>
CC: Greg Smith <gsmith@gregsmith.com>, Mark Kirkwood <markir@paradise.net.nz>,
	"pgsql-performance@postgresql.org" <pgsql-performance@postgresql.org>
Date: Thu, 2 Apr 2009 13:20:15 -0700
Subject: Re: Raid 10 chunksize
Thread-Topic: [PERFORM] Raid 10 chunksize
Thread-Index: AcmzvK25NfeDmBgCQG+VKwO5ZOYzegAE8ENv
Message-ID: <C5FA6C0F.41E8%scott@richrelevance.com>
In-Reply-To: <b42b73150904021058v2b390b04k53aeabc58eb4e769@mail.gmail.com>
Accept-Language: en-US
Content-Language: en
acceptlanguage: en-US
Content-Type: text/plain; charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
MIME-Version: 1.0


On 4/2/09 10:58 AM, "Merlin Moncure" <mmoncure@gmail.com> wrote:

> On Wed, Mar 25, 2009 at 12:16 PM, Scott Carey <scott@richrelevance.com> w=
rote:
>> On 3/25/09 1:07 AM, "Greg Smith" <gsmith@gregsmith.com> wrote:
>>> On Wed, 25 Mar 2009, Mark Kirkwood wrote:
>>>> I'm thinking that the raid chunksize may well be the issue.
>>>=20
>>> Why? =A0I'm not saying you're wrong, I just don't see why that paramete=
r
>>> jumped out as a likely cause here.
>>>=20
>>=20
>> If postgres is random reading or writing at 8k block size, and the raid
>> array is set with 4k block size, then every 8k random i/o will create TW=
O
>> disk seeks since it gets split to two disks. =A0 Effectively, iops will =
be cut
>> in half.
>=20
> I disagree.  The 4k raid chunks are likely to be grouped together on
> disk and read sequentially.  This will only give two seeks in special
> cases. =20

By definition, adjacent raid blocks in a stripe are on different disks.


> Now, if the PostgreSQL block size is _smaller_ than the raid
> chunk size,  random writes can get expensive (especially for raid 5)
> because the raid chunk has to be fully read in and written back out.
> But this is mainly a theoretical problem I think.

This is false and a RAID-5 myth.  New parity can be constructed from the ol=
d
parity + the change in data.  Only 2 blocks have to be accessed, not the
whole stripe.

Plus, this was about RAID 10 or 0 where parity does not apply.

>=20
> I'm going to go out on a limb and say that for block sizes that are
> within one or two 'powers of two' of each other, it doesn't matter a
> whole lot.  SSDs might be different, because of the 'erase' block
> which might be 128k, but I bet this is dealt with in such a fashion
> that you wouldn't really notice it when dealing with different block
> sizes in pg.

Well, raid block size can be significantly larger than postgres or file
system block size and the performance of random reads / writes won't get
worse with larger block sizes.  This holds only for RAID 0 (or 10), parity
is the ONLY thing that makes larger block sizes bad since there is a
read-modify-write type operation on something the size of one block.

Raid block sizes smaller than the postgres block is always bad and
multiplies random i/o.

Read a 8k postgres block in a 8MB md raid 0 block, and you read 8k from one
disk.
Read a 8k postgres block on a md raid 0 with 4k blocks, and you read 4k fro=
m
two disks.