Date: Sat, 30 Mar 1996 16:27:38 -0800 (PST) From: Tom Samplonius <tom@uniserve.com> To: "Marc G. Fournier" <scrappy@ki.net> Cc: stable@FreeBSD.org Subject: Re: SCSI Performance under -stable/-current Message-ID: <Pine.BSF.3.91.960330162155.29246A-100000@haven.uniserve.com> In-Reply-To: <Pine.BSF.3.91.960330161449.3268C-100000@ki.net>
next in thread | previous in thread | raw e-mail | index | archive | help
On Sat, 30 Mar 1996, Marc G. Fournier wrote: > I'm also sending this to -current, in case there are bugs in > -stable that are fixed in -current, that might be reflected here :) > > My results show: > > ncr:sd0 - 1.95write/3.01read - Conner CFP1060S > ncr:sd1 - 1.75write/2.50read - Conner CFA540S > aha:sd2 - 1.06write/1.06read - Quantum LPS340S > aha:sd3 - 1.12write/0.94read - Quantum LP240S > > ncr == ASUS SC-200 PCI Controller > aha == Adaptec 1542CF ISA Controller > > Do those values from IOzone look right? Looking at the aha:sd? > values, they look low to me, especially considering that when I ran IOzone > on it, there would have been zero activity on those drives except for > their respective swap partitions: The aha results seem alright. Remember the 1542 is only a 16bit controller. The ncr results should be higher, unless your drives are slow. On a -stable system with a 2940 and a Quantum Fireball, I see over 6MB/sec on file I/O. Tom
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?Pine.BSF.3.91.960330162155.29246A-100000>