lkml.org 
[lkml]   [2001]   [Aug]   [24]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: 3ware RAID1 sequential read speed slower than write speed (2.4.8-ac10)
On Fri, Aug 24, 2001 at 09:06:55AM -0000, Yusuf Goolamabbas wrote:
> P3-450 with 256MB RAM with a 3ware 6200 attached to 2 20GB Western
> Digitial 7200RPM Caviar Drive WD200BB
>
> Running bonnie++ <http://www.coker.com.au/bonnie++/> on both an ext3
> and ext2 partition, I get the following results
>
> Version 1.01d ------Sequential Output------ --Sequential Input- --Random-
> -Per Chr- --Block-- -Rewrite- -Per Chr- --Block-- --Seeks--
> Machine Size K/sec %CP K/sec %CP K/sec %CP K/sec %CP K/sec %CP /sec %CP
> ext3 512M 6020 98 41022 57 6673 8 4619 77 21834 14 237.1 2
>
>
> Version 1.01d ------Sequential Output------ --Sequential Input- --Random-
> -Per Chr- --Block-- -Rewrite- -Per Chr- --Block-- --Seeks--
> Machine Size K/sec %CP K/sec %CP K/sec %CP K/sec %CP K/sec %CP /sec %CP
> ext2 512M 6385 99 39969 32 9371 10 5540 91 27864 14 320.7 2
>
>
> Whilst, I think the write performance is quite good. The read
> performance seems to be quite bad. I expected read performance to be
> better than write performance for RAID-1 configuration
>
> Any ideas,patches to try ?

Yes, for a mirrored setup you would expect to see almost double the random
seeks and read rates equaling write rates. I think the block size that
bonnie uses is small, so it doesn't give the RAID a chance to alternate
disks for block reads.

At least this is what seemed to happen when I was testing software RAID-1
using bonnie (not bonnie++).

What's the speed of one disk? Can you try 2.4.8-ac9, -ac10 includes a 3ware
driver update.

I'll post some numbers from the software raid testing I was doing when I get
to work for comparison. If anyone wants to send me a 3ware card (hint,
hint ;-), I'll gladly run some tests on that as well to compare. Heck, I'll
even write a full review comparing 3ware to software raid!

-Dave
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/

\
 
 \ /
  Last update: 2005-03-22 12:58    [W:0.024 / U:0.016 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site