Subject: Re: RaidFrame poor performance
To: None <current-users@NetBSD.org>
From: Jochen Keil <J.Keil@gmx.de>
Date: 01/28/2005 12:01:16
Matthias Scheler wrote:
> In article <41F7C157.email@example.com>,
> Jochen Keil <J.Keil@gmx.de> writes:
>>The iozone benchmark are very interesting (at least for me). A 3 disk R5
>>setup (with two usable disks) set up with a correct stripe/fs block size
>>is about 4 times faster than a 4 disk R5 (3 usable disks) setup.
> What's surprising about that? On a 3 disk RAID 5 you need 3 reads and
> 3 writes for 1 write to the RAID. On a 4 disk RAID 5 you need 4 of them.
This is why i wrote "(at least for me)". Anyway a 400% performance
increase is quite remarkable in my opinion. Whatsmore with a random
write speed of ~44mb/s the R5 set is almost as fast a single disk
> RAID 5 in hardware is slow. RAID 5 in software is very, very slow.
> This is not specific to RAIDframe but to every RAID 5 software
(I'm replying here if you don't mind)
As you can see from my benchmark results your statement is not correct.
I for myself think that 44mb/s is rather good and Gregs SCSI R5 is just