Date: Thu, 17 Feb 2000 11:43:19 -0500 From: Mike Tancsa <mike@sentex.ca> To: current@freebsd.org Cc: msmith@freebsd.org Subject: MegaRAID 428 vs 466 and a way to wedge the controller Message-ID: <3.0.5.32.20000217114319.01bd76a0@marble.sentex.ca>
next in thread | raw e-mail | index | archive | help
OK,
I seem to have my 466 (aka Perc 2/SC) working somewhat as expected once I
flashed to BIOS to the latest rev from AMI. However, I can now more
reliably wedge the controller where as with the 428 this would only happen
on occasion.
I have a 466 with a RAID5 set on 3 quantum LVD drives. If I do 4 copies of
bonnie -s 600 -d /mnt &
the controller seems to wedge on a fairly regular basis, where as this only
happened to me once with the 428. As for performance, I do see good
performance on random reads, but not so great performance on raw io. e.g.
-------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
amrd0 500 5451 22.7 5275 6.5 4756 6.8 12879 80.6 28521 20.9 840.0
6.1
ad0 500 17559 72.3 18083 22.8 5060 9.6 15293 96.2 14258 15.3 426.4
4.0
ad0: 13072MB <QUANTUM FIREBALLP KX13.6> [26559/16/63] at ata0-master using
UDMA33
This is with a 64K stripe. I imagine a 128K stripe will drop the random
seeks, but improve the throughput ? Eventually, this will be a pop3 server
so I am not quite sure yet how best to optimze for that.
---Mike
------------------------------------------------------------------------
Mike Tancsa, tel +1 519 651 3400
Network Administrator, mike@sentex.net
Sentex Communications www.sentex.net
Cambridge, Ontario Canada
To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-current" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?3.0.5.32.20000217114319.01bd76a0>