rbthomas@JOVE.RUTGERS.EDU (10/31/90)
> > Think we should apply for > a grant of some Cray time to run clockres on it? I'm not that curious. > > Marion > I was (curious, that is), so I ran it on the Y/MP at PSC. This is what I got. Script started on Tue Oct 30 23:19:03 1990 pscymp$ for i in 1 2 3 4 5 6 7 8 9 0 > do > clockres > done Clock step resolution 29 microseconds or better Clock step resolution 30 microseconds or better Clock step resolution 32 microseconds or better Clock step resolution 29 microseconds or better Clock step resolution 32 microseconds or better Clock step resolution 33 microseconds or better Clock step resolution 30 microseconds or better Clock step resolution 30 microseconds or better Clock step resolution 30 microseconds or better Clock step resolution 30 microseconds or better pscymp$ <EOF> <EOF> Script done on Tue Oct 30 23:19:23 1990 For comparison purposes, I ran it on a Convex C1/XP we have locally, and got this. Does this mean that the Y/MP is only 30 percent faster than the C1/XP ? (;->) Script started on Tue Oct 30 23:27:32 1990 kingtut$ for i in 1 2 3 4 5 6 7 8 9 0 > do > clockres > done Clock step resolution 36 microseconds or better Clock step resolution 38 microseconds or better Clock step resolution 39 microseconds or better Clock step resolution 36 microseconds or better Clock step resolution 36 microseconds or better Clock step resolution 36 microseconds or better Clock step resolution 38 microseconds or better Clock step resolution 38 microseconds or better Clock step resolution 36 microseconds or better Clock step resolution 38 microseconds or better kingtut$ ^D script done on Tue Oct 30 23:28:08 1990 ======================================================================== Rick Thomas, Manager Supercomputer Remote Access Center, Rm D117 Rutgers University, College of Engineering Brett and Bowser Roads Piscataway, NJ 08855-0909 Phone: (201) 932-4301 Internet: rbthomas@jove.rutgers.edu UUCP: {any backbone site}!rutgers!rbthomas Alternate UUCP: {convex|c1apple|karna}!kingtut!rbthomas "Soap and education are not as sudden as a massacre, but they are more deadly in the long run." -- Mark Twain ========================================================================
Mills@udel.edu (10/31/90)
Rick, I have reports that the Cray can keep time to the basic CPU cycle, something less than 10 ns. If that really is the case, NTP can locate your CPU within the machine room to less than ten feet. Dave