|
|
|
|
|
|
|
|
|
|
|
|
| home | help | back | first | fref | pref | prev | next | nref | lref | last | post |
To: Wallace Colyer <wally+@andrew.cmu.edu> Cc: Info-AFS@transarc.com In-Reply-To: Your message of "Tue, 24 Sep 91 16:28:23 EDT." Date: Tue, 24 Sep 91 14:53:53 -0700 From: "Donald R. Proctor (510/987-0356)" <spgdrp@ganges.ucop.edu> > We have noticed for quite sometime that DECStations are considerably > slower than Sparcstations in dealing with AFS (as clients). So much > slower that better IO and/or faster machines don't seem to count for the > differences. Has anyone noticed this, or done any performance studies? > > -Wallace This might be because of the way Ultrix limits the amount of memory available for caching. You're supposed to be able to get around this by tinkering with the -chunk parameter when invoking afsd, but this has never worked properly for us. As a consequence, we use disk caching, which is intuitively somewhat slower than memory caching. --Don
|
|
|
|
|
|
|
|
|
|
|
|
| home | help | back | first | fref | pref | prev | next | nref | lref | last | post |