Date: Mon, 6 Nov 2000 17:22:10 +0100 (CET) From: Mads Bondo Dydensborg <madsdyd@challenge.dk> To: Doug Ledford <dledford@redhat.com> Cc: "'aic7xxx@FreeBSD.ORG'" <aic7xxx@FreeBSD.ORG> Subject: Re: Question about throughput Message-ID: <Pine.LNX.4.21.0011061716370.2261-100000@challenge.dk> In-Reply-To: <3A071DDD.B611DB5D@redhat.com>
next in thread | previous in thread | raw e-mail | index | archive | help
On Mon, 6 Nov 2000, Doug Ledford wrote: > Mads Bondo Dydensborg wrote: > > I have tried to analyze what happens, and > > it seems that the aic7xxx driver will only issue a single read command at > > a time. When writing, multiple write commands will be issued. > > The driver will issue as many reads to the device as the mid level and higher > level code sends to it. Under normal conditions, large numbers of reads are > queued by the block device read-ahead code in the upper block layers, not by > the SCSI subsystem. So, in order for linux to do the same thing for your > device, you would have to be accessing it through the block device layers > (aka, your drive appears as any other normal disk drive and you aren't using a > raw device to access it) and you would have to be doing sequential reads. If > you meet those two items, then linux should be using the tagged queueing to > access your drive with multiple reads outstanding. I do meet those two items. I assume tagged queueing to be the case then. > > > Obviously, using the SCI network is not the same as the local case. There > > is a greater latency for operations. > > > > My question is this; are there any way to force the aic7xxx driver to > > issue more read commands to a device? > > > > Any help, hints, suggestions, comments, flames are welcome. Anything, > > really. :-) I must defend my results in 3 weeks, and would _really_ like > > to have solved this problem first. > > I would try making a filesystem on the SCSI network device, then mounting that > filesystem, then run the bonnie program on that filesystem to see how it > performs under those conditions. I assume that's not what you are doing now > ;-) At the moment I am using dd to test the performance on a md raid filesystem. I will try bonnie. (Add. info: In my thesis I used a slower disk, and have tested that with both dd and bonnie. I now have access to faster disks - the assymmetrical problem only takes place on these faster disks. Very painful). > Using bonnie on a mounted filesystem on a local disk usually results in > sequential reads being better than writes until writes eclipse the 85MB/s > mark, at which point they start to pass reads (on a dual PII-500 I think it > was, and the reads were running out of CPU power for searching through the > buffer cache before each read command was issued as I recall). Interessting. I will try experimenting with the amount of physical ram. Due to some limitations in SCI, I will never be able to go above 132/2 = 66 MB/s however. Thanks a lot for your response and help (so far! :-) Mads -- Mads Bondo Dydensborg. madsdyd@challenge.dk Did you ever notice how at trade shows Microsoft is always the one giving away stress balls... To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe aic7xxx" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?Pine.LNX.4.21.0011061716370.2261-100000>