Date: 15 Oct 1999 19:14:15 +0000 From: Randell Jesup <rjesup@wgate.com> To: scsi@FreeBSD.ORG, Michael@sinz.org (Mike Sinz) Subject: Re: FreeBSD 3.2 / Slow SCSI Dell PowerEdge 4300 Message-ID: <ybu6708wi60.fsf@jesup.eng.tvol.net.jesup.eng.tvol.net> In-Reply-To: "Kenneth D. Merry"'s message of "Fri, 15 Oct 1999 14:17:41 -0600 (MDT)" References: <199910152017.OAA51280@panzer.kdm.org>
next in thread | previous in thread | raw e-mail | index | archive | help
"Kenneth D. Merry" <ken@kdm.org> writes:
>> (based on bonnie results for individual disks with WCE as shipped (off i think)
>> you get better sequential performance and more importantly for most of us
>> better seeking)
>
>I'd like to see some hard numbers for these before I change the quirk. (i.e.
>bonnie numbers at 1, 2, 4, 8, 16, 32, and 64 tags, with and without write
>caching) camcontrol can be used to change the tags and enable/disable
>write caching on the fly. If anyone wants to do this, see PR kern/10398
>for an example of what sort of tests to run.
Knowing SCSI and how drives and FS's work, there might really be a
benefit in a multi-tasking OS (especially with multiple filesystems) to
running at least a small number of tags even at the cost of 5% on
sequential writes (which is the order that we're talking of). There's
probably no need for 32 or 64, but 4 or 8, maybe 16 on this drive might
be reasonable.
Looking at the bonnie results from 10398:
write cache enabled
-------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
Number of Tags
NO 100 7222 89.2 6801 21.8 2347 8.5 7330 93.3 7368 14.6 226.5 4.7
2 100 7263 90.3 6357 20.3 2730 9.9 7025 90.5 7321 14.9 209.4 4.6
3 100 7115 88.1 6406 20.8 2289 8.9 7307 93.9 7335 15.0 212.6 4.5
4 100 7281 90.0 6204 20.8 2278 9.1 7267 93.7 7350 15.3 217.6 4.8
8 100 7236 89.7 6007 19.4 2284 8.7 7239 93.1 7374 14.9 213.4 4.5
16 100 6775 83.7 6110 19.5 2283 8.7 7239 93.1 7380 14.8 217.8 4.8
32 100 7265 89.9 4385 13.9 2274 8.7 7302 93.7 7324 14.5 218.9 4.5
64 100 6731 83.3 3038 9.8 2271 8.7 7337 94.6 7356 14.7 213.6 4.4
Character sequential output is basically unaffected by tags
(bounces around a lot). Block output goes down; one jump from NO to 2 of
450K (but that's suspect, since it goes up 50K for 3); overall after the
initial drop it goes down circa 20K-ish per tag through 16 tags; it gets
worse above 16. Rewrite is unaffected. Input is unaffected. Seeks are
apparently unaffected (there's a small drop, but it may well be within
measurement error - certainly there's no continuing drop).
Perhaps a more interesting test would be a wall-time comparison
with a make world or some such - lots of writing and reading in a
real-world sort of situation.
Good multitasking disk benchmarks are hard to come by.
--
Randell Jesup, Worldgate Communications, ex-Scala, ex-Amiga OS team ('88-94)
rjesup@wgate.com
To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-scsi" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?ybu6708wi60.fsf>