David Mansfield wrote:
>
> I am concerned that these stats for bonnie are in fact pretty poor for the
> hardware involved. Could someone running SW RAID5 on similar hardware
> comment on your bonnie findings?
I missed the original message unfortunately, so
I don't know what the original hardware described
is. However, I've just done some ``interesting''
tests comparing software and hardware RAID on similar
Linux servers. Using bonnie, the software RAID
was basically 3-4 times faster. Here are the
exact stats:
Software RAID: Timing buffered disk reads: 32 MB in 0.93 seconds =34.41 MB/sec
Hardware RAID: Timing buffered disk reads: 32 MB in 2.42 seconds =13.22 MB/sec
-------Sequential Output-------- ---Sequential Input-- --Random--
-Per Char- --Block--- -Rewrite-- -Per Char- --Block--- --Seeks---
Machine MB K/sec %CPU K/sec %CPU K/sec %CPU K/sec %CPU K/sec %CPU /sec %CPU
software 1024 7634 87.3 34964 37.3 12468 28.6 8735 95.7 31303 28.7 172.2 4.2
hardware 2000 6786 76.2 8487 6.5 4907 11.4 8205 92.2 12943 18.9 227.8 6.9
and the hardware specs:
Software RAID Hardware RAID
1*Intel P-II 450 Quad Intel P-II 450
m/b SuperMicro P6DBE Unknown: Dell Poweredge motherboard?
mem 256M 1G
scsi 2*Advansys 940UW
raid AMI MegaRAID
drives 6*Barracuda 18GB 6*Barracuda 18GB
config RAID-5 RAID-5
[Note: I think the fact that the hardware RAID had 4
processors and 4 times as much RAM shouldn't actually
make too much difference to the benchmarks].
Anyhow, a bit of a victory for software RAID on Linux.
Rich.
--
Bibliotech Ltd., 631-633 Fulham Rd., London SW6 5UQ. +44 171 460 4646
http://www.bibliotech.co.uk/ ------ Home page: http://www.annexia.org/
Work email: [EMAIL PROTECTED] ----- Home: [EMAIL PROTECTED]
Original message content Copyright © 1999 Richard Jones
