From mboxrd@z Thu Jan  1 00:00:00 1970
From: Roberto Spadim <roberto@spadim.com.br>
Subject: Re: write-behind has no measurable effect?
Date: Tue, 15 Feb 2011 11:26:46 -0200
Message-ID: <AANLkTimdyU-rQMY83fq9r2LpP0PGgz0gfV+=gkQWH++F@mail.gmail.com>
References: <20110214213817.GG836@hellgate.intra.guy>
	<20110215095042.51ef7e0a@notabene.brown>
	<20110214225754.GK19990@hellgate.intra.guy>
	<20110215104109.06b12b33@notabene.brown>
	<20110215010052.GA13135@hellgate.intra.guy>
	<4D59D4A5.9050106@anonymous.org.uk>
	<20110215021900.GB13135@hellgate.intra.guy>
	<AANLkTikFSOePZJXknAt=Tx6+FpdJ4tiSNwpuwuPC3RY=@mail.gmail.com>
	<AANLkTinAV4jJc8ueZ1mjRPpDnA_eWhUDcL8uz-0Ufifk@mail.gmail.com>
	<20110215124026.GC13135@hellgate.intra.guy>
Mime-Version: 1.0
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: QUOTED-PRINTABLE
Return-path: <linux-raid-owner@vger.kernel.org>
In-Reply-To: <20110215124026.GC13135@hellgate.intra.guy>
Sender: linux-raid-owner@vger.kernel.org
To: Andras Korn <korn@raidlist.elan.rulez.org>
Cc: linux-raid@vger.kernel.org
List-Id: linux-raid.ids

writes will be the speed of slowest mirror (ssd or hd)
read can have speed improvement with this patch

there=B4s some options

/sys/block/md0/md/read_balance_mode

near_head =3D today implementation
round_robin =3D> could be usefull if you have ssd only, since
round_robin consider that access time is the same for any drive (hard
disk access time is diferent for random and sequencial)
stripe =3D> i didn=B4t get good benchmarks, but it=B4s nice to have it,
since we could put it over network, some sectors on one disk others on
another (nbd)
time_based =3D> here you should send me some information about you disk
for example:
    access time of mirrors (check your drive information website, ssd
are normaly <0.1ms, hard disks near 10~20 ms)
    sequencial read speed (use dd if=3D/dev/sda of=3D/dev/zero bs=3D409=
6,
change the block size to size you will use with your filesystem, for
vertex2 ssd i=B4m using bs=3D4096, for disks it=B4s a good value too, s=
ince
disks bs ~=3D number of heads (2,4,8) )

send me the access time and sequencial read speed and i make the
values to tune your sysfs (/sys/block/md0/md/read_balance_config)

you will need to :

echo "time_based" > /sys/block/md0/md/read_balance_mode
echo "disks informations" > /sys/block/md0/md/read_balance_config

for each mirror, you can=B4t use sysfs file to configure it, maybe a
bash script is a better solution to configure it, on future version i
will change it and put at /sys/block/md0/md/dev-xxxx/

2011/2/15 Andras Korn <korn@raidlist.elan.rulez.org>:
> On Tue, Feb 15, 2011 at 06:10:17AM -0300, Roberto Spadim wrote:
>
>> andras could you make some benchmarks to raid1 with round robin read=
 balance?
>> at this site:
>> www.spadim.com.br/raid1/
>>
>> it's kernel 2.6.37 based
>
> Yes, I can do that. Can you give me some hints on what specific
> configuration to try? I see you have some sysfs tunables. My raid1 ar=
ray
> consists of two spinning disks and an SSD, all local.
>
> Do you expect this patch to make a difference in my case? With the sp=
inning
> disks marked as write-mostly, I'm getting close to the read performan=
ce of
> the SSD (except for very small random reads, for some reason).
>
> It's random writes that are much slower than with only the SSD.
>
> --
> =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 Andras Korn <korn at elan.rul=
ez.org>
> =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0There is no spoon(). But there is =
a fork().
> --
> To unsubscribe from this list: send the line "unsubscribe linux-raid"=
 in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at =A0http://vger.kernel.org/majordomo-info.html
>



--=20
Roberto Spadim
Spadim Technology / SPAEmpresarial
--
To unsubscribe from this list: send the line "unsubscribe linux-raid" i=
n
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html