Date: Tue, 15 Feb 2000 12:27:14 +1030 From: Greg Lehey <grog@lemis.com> To: "Karsten W. Rohrbach" <karsten@rohrbach.de> Cc: Christoph Kukulies <kuku@gilberto.physik.RWTH-Aachen.DE>, hackers@FreeBSD.ORG, shimon@simon-shapiro.org Subject: Re: bonnie still trustable? Message-ID: <20000215122714.D4538@freebie.lemis.com> In-Reply-To: <20000215023758.A99113@rohrbach.de> References: <200002110949.KAA14176@gil.physik.rwth-aachen.de> <20000213151650.A97412@freebie.lemis.com> <20000215023758.A99113@rohrbach.de>
next in thread | previous in thread | raw e-mail | index | archive | help
On Tuesday, 15 February 2000 at 2:37:58 +0100, Karsten W. Rohrbach wrote: > Greg Lehey(grog@lemis.com)@Sun, Feb 13, 2000 at 03:16:50PM +1030: >> On Friday, 11 February 2000 at 10:49:24 +0100, Christoph Kukulies wrote: >>> PIII/500, 128 MB >>> >>> I'm wondering if this is trustable: >>> >>>> bonnie -s 400 >>> File './Bonnie.14321', size: 419430400 >>> Writing with putc()...done >>> Rewriting...done >>> Writing intelligently...done >>> Reading with getc()...done >>> Reading intelligently...done >>> Seeker 1...Seeker 2...Seeker 3...start 'em...done...done...done... >>> -------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 >>> 400 20015 73.7 18369 22.9 6750 12.6 22308 81.5 22467 26.0 93.8 1.0 >>> ^^^^ >>> ? >>> >>> >>> wdc0 at 0x1f0-0x1f7 irq 14 flags 0xa0ff on isa >>> wdc0: unit 0 (wd0): <IBM-DPTA-372050>, DMA, 32-bit, multi-block-16 >>> wd0: 19574MB (40088160 sectors), 39770 cyls, 16 heads, 63 S/T, 512 B/S >> >> There are lies, damn lies, and statistics. I'm sure that the results >> mean something; the real question is, what do you want them to mean? >> >> If you're trying to measure the storage device, rawio (Ports >> Collection) is a much better choice. > > i personally like shimon's st.d... > it exercises a disk very thoroughly, so if you want to benchmark - and, > of course, have plenty of time for the box to run the benchmark - st.d > is the choice since you got your free-of-charge(tm) burn in of the disk > subsystem with it. what it basically does is simulating a nonlinear load > behaviour like you got in a multi-client or large amount of concurrent > i/o processes environment and i like it a lot. welps, i won't argue > about the command line parameters - just see for yourself ;-) > > http://www.simon-shapiro.org/st_d/index.html Interesting. This looks like a well-kept secret. I don't see anything that jumps out and tells me where to get this software, nor where I can get a printable version of the documentation. Shimon, can you help? Greg -- Finger grog@lemis.com for PGP public key See complete headers for address and phone numbers 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?20000215122714.D4538>