From: Yuehai Xu <yuehaixu@gmail.com>
To: linux-raid@vger.kernel.org
Cc: gombasg@sztaki.hu, yhxu@wayne.edu
Subject: Re: The huge different performance of sequential read between RAID0 and RAID5
Date: Thu, 28 Jan 2010 09:31:23 -0500 [thread overview]
Message-ID: <100eff551001280631h7de01ba6n52d79fdfcea9445e@mail.gmail.com> (raw)
In-Reply-To: <20100128070606.GD3098@boogie.lpds.sztaki.hu>
On Thu, Jan 28, 2010 at 2:06 AM, Gabor Gombas <gombasg@sztaki.hu> wrote:
> On Wed, Jan 27, 2010 at 10:16:12PM -0500, Yuehai Xu wrote:
>
>> md0 : active raid5 sdh1[7] sdg1[5] sdf1[4] sde1[3] sdd1[2] sdc1[1] sdb1[0]
>> 631353600 blocks level 5, 64k chunk, algorithm 2 [7/6] [UUUUUU_]
> [...]
Do you mean there is something wrong when I setup my RAID5? The
command I use to setup RAID5 is:
mdadm --create /dev/md0 --level=5 --raid-devices=7 /dev/sdb1 /dev/sdc1
/dev/sdd1 /dev/sde1 /dev/sdf1 /dev/sdg1 /dev/sdh1
I don't think any of my drive fail because there is no "F" in my
/proc/mdstat output
>> Then I start IOZONE which starts 10 processes to do the sequential
>> read(iozone -i 1). Each process read 640M file on each partition. The
>> throughput of RAID0 is about 180M/s, while the throughput of RAID5 is
>> just 43M/s. Why the performance between RAID0 and RAID5 is so
>> different?
>
> You have a degraded RAID5 array with one drive missing, meaning the data
> has to be recalculated from parity all the time. That obviously kills
> performance.
>
> Gabor
How do you know my RAID5 array has one drive missing? I tried to setup
RAID5 with 5 disks, 3 disks, after each setup, recovery has always
been done. However, if I format my md0 with such command:
mkfs.ext3 -b 4096 -E stride=16 -E stripe-width=*** /dev/XXXX, the
performance for RAID5 becomes usual, at about 200~300M/s.
>
> --
> ---------------------------------------------------------
> MTA SZTAKI Computer and Automation Research Institute
> Hungarian Academy of Sciences
> ---------------------------------------------------------
>
--
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 http://vger.kernel.org/majordomo-info.html
next prev parent reply other threads:[~2010-01-28 14:31 UTC|newest]
Thread overview: 9+ messages / expand[flat|nested] mbox.gz Atom feed top
2010-01-28 3:16 The huge different performance of sequential read between RAID0 and RAID5 Yuehai Xu
2010-01-28 7:06 ` Gabor Gombas
2010-01-28 14:31 ` Yuehai Xu [this message]
2010-01-28 14:41 ` Gabor Gombas
2010-01-28 14:55 ` Yuehai Xu
2010-01-28 15:27 ` Robin Hill
2010-01-29 6:05 ` Michael Evans
2010-01-29 11:53 ` Goswin von Brederlow
2010-01-30 7:03 ` Michael Evans
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=100eff551001280631h7de01ba6n52d79fdfcea9445e@mail.gmail.com \
--to=yuehaixu@gmail.com \
--cc=gombasg@sztaki.hu \
--cc=linux-raid@vger.kernel.org \
--cc=yhxu@wayne.edu \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).