On Sun, 8 Nov 1998, Chris J. Magnuson wrote:
> Here are some stats on a 16-bit fast wide single-channel ICP Vortex PCI
> RAID-5 card on a 3-drive array.
>
> The drives are Fujitsu 9.1 Gig narrow M2949 drives (14.x Meg/sec
> theoretical limit on speed). The system is a 233MHz Supermicro Pentium MMX
How do fast narrow drives (10MB/s SCSI) theoretically do 14.x Meg/sec?
> on Redhat Linux 5.1, with 128 Megs of RAM. When I ran these numbers with
> bonnie -200 I was running X11 and some other fairly lightweight processes
>
> -------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
> 200 2419 34.3 2829 17.7 1865 21.2 3821 41.8 5767 27.2 148.0 6.8
>
> Comments on these numbers welcome, I just briefly scanned the bonnie web
Those numbers blow. I'm really hoping when I replace one of our servers
with a PII and hardware RAID5, that I can not only increase reliability,
but also increase disk i/o speed. Someone from ICP recently posted this:
RAID5
-------Sequential Output-------- ---Sequential Input-- --Random--
-Per Char- --Block--- -Rewrite-- -Per Char- --Block--- --Seeks---
MB K/sec %CPU K/sec %CPU K/sec %CPU K/sec %CPU K/sec %CPU /sec %CPU
750 14465 80.1 6305 5.7 6938 14.6 14785 61.8 29818 19.1 149.7 2.4
That was on a 5 drive array spread across 3 channels. They had 32mb 50ns
cache too. Those are the kinds of numbers I want. Can someone from ICP
comment on why Chris's setup blows so badly in comparison? I really doubt
his drives are the bottleneck.
---dont't waste your cpu, crack rc5...www.distributed.net team enzo---
Jon Lewis <[EMAIL PROTECTED]> | Spammers will be winnuked or
Network Administrator | nestea'd...whatever it takes
Florida Digital Turnpike | to get the job done.
______http://inorganic5.fdt.net/~jlewis/pgp for PGP public key________
