Date: Wed, 11 Nov 1998 22:47:09 +0100 (MET) From: Mats Lofkvist <mal@algonet.se> To: hackers@FreeBSD.ORG Subject: Re: SCSI vs. DMA33.. Message-ID: <199811112147.WAA10051@kairos.algonet.se> References: <199811111842.KAA18548@hub.freebsd.org>
next in thread | previous in thread | raw e-mail | index | archive | help
> I must say, I'm surprised. This makes it look like there's more of a
> performance hit with concurrent requests on SCSI than on IDE. Let's
> look at the performance hits as a percentage (there's some guesswork
> which is which, of course, but only a little):
>
> alone 2 together % drop
> wd2 10.2 8.8 14
> wd3 10.0 8.8 12
> da0 5.7 3.3 42
> da1 3.7 2.7 23
Don't know what conclusions to draw, but I don't see this decrease
in performance with concurrent requests. Current-smp on a providence
mb (7880 on board) with a quantum viking 4.5GB uw (da0) and a vikingII
4.5GB uw (da1):
bash# dd if=/dev/rda0c bs=32k count=10000 of=/dev/null
10000+0 records in
10000+0 records out
327680000 bytes transferred in 30.950317 secs (10587291 bytes/sec)
bash# dd if=/dev/rda1c bs=32k count=10000 of=/dev/null10000+0 records in
10000+0 records out
327680000 bytes transferred in 23.328261 secs (14046482 bytes/sec)
bash# dd if=/dev/rda0c bs=32k count=10000 of=/dev/null & dd if=/dev/rda1c bs=32k count=13000 of=/dev/null
[1] 375
13000+0 records in13000+0 records out
425984000 bytes transferred in 30.822861 secs (13820391 bytes/sec)
bash# 10000+0 records in
10000+0 records out
327680000 bytes transferred in 31.760830 secs (10317111 bytes/sec)
alone 2 together % drop
da0 10.6 10.3 1.6
da1 14.0 13.8 2.6
_
Mats Lofkvist
mal@algonet.se
To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-hackers" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199811112147.WAA10051>