Date: Fri, 18 Oct 1996 00:22:11 -0400 (EDT) From: "Marc G. Fournier" <scrappy@ki.net> To: Michael Hancock <michaelh@cet.co.jp> Cc: Satoshi Asami <asami@FreeBSD.org>, current@FreeBSD.org Subject: Re: Iozone: local vs nfs drives Message-ID: <Pine.NEB.3.95.961018002044.6698D-100000@quagmire.ki.net> In-Reply-To: <Pine.SV4.3.93.961018112301.29655B-100000@parkplace.cet.co.jp>
next in thread | previous in thread | raw e-mail | index | archive | help
On Fri, 18 Oct 1996, Michael Hancock wrote: > On Thu, 17 Oct 1996, Satoshi Asami wrote: > > > * Did you have tagged-command queuing enabled? According to my tests, > > * that will make a BIG difference, if you have decent drives. > > > > I haven't seen tagged-queueing make much difference for reads. These > > are the kind of numbers we've seen before (off the top of my head): > > > > w/o tag with tag > > W R W R > > Quantum Atlas 6 7 6 7 > > Seagate 'Cuda 4 6 6 7 > > I guess we haven't addressed the original question though. No, it doesn't, but I think it might bring up another point... is there a way of having tagged-command queueing enabled by default? If there is the potential to see such a boost in performance as has been reported, at least in the 'write' department...is there a disadvantage to it that hasn't been brought up? Marc G. Fournier scrappy@ki.net Systems Administrator @ ki.net scrappy@freebsd.org
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?Pine.NEB.3.95.961018002044.6698D-100000>