control for one variable. what does 10.6.8 with 1.6.0 do? does scan stat() a lot of stuff? no bulkstat, currently.
On Wed, Oct 26, 2011 at 10:13 AM, Ken Hornstein <[email protected]> wrote: > We've recently got some MacOS Lion systems in, and we've noticed that OpenAFS > seems to perform rather poorly on then. > > Specifically, we've noticed this when using nmh on these systems. Here's > an example of what I mean. On an older iMac running 10.6.8/OpenAFS 1.4.11: > > % /usr/bin/time scan +kerberos > 4.29 real 0.02 user 0.08 sys > > On a brand-new new Mac Pro system running Lion/OpenAFS 1.6.0 > % /usr/bin/time scan +kerberos > 44.29 real (I don't remember the rest, it was on another system) > > These systems are on the exact same network, so that's not the issue. > > I realize that even the "old" Mac performance isn't wonderful. On my > end-of-lifed-12-years-ago Sparc system running Solaris 10 (same deal, > on the same exact network): > > % /usr/bin/time scan +kerberos > real 1.6 > user 0.1 > sys 0.5 > > Any ideas where to start tracking this down? I could maybe believe > something like statahead is the issue; the scan performance on the Lion > system is very "spurty". > > --Ken > _______________________________________________ > OpenAFS-info mailing list > [email protected] > https://lists.openafs.org/mailman/listinfo/openafs-info > -- Derrick _______________________________________________ OpenAFS-info mailing list [email protected] https://lists.openafs.org/mailman/listinfo/openafs-info
