[Beowulf] LSF and cpu factor
Many of your questions may have already been answered in earlier discussions or in the FAQ. The search results page will indicate current discussions as well as past list serves, articles, and papers.
Mark Hahn hahn at physics.mcmaster.caThu Aug 25 05:56:44 PDT 2005
- Previous message: [Beowulf] LSF and cpu factor
- Next message: [Beowulf] lsf mpi job run policy !!!
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
> i need the correct value of the cpu factor relative to default value (1) > in LSF. "lsinfo -M" command lists the predefined architectures and their > properties but i couldn't information about following architectures > (output of the "lim -t" command), > > SUNWSunFireV210_1002_sparcv9 : sunfire v210 2 cpu cluster > SUNWSunFire15000_900_sparcv9 : sunfire12k smp machine why not just look up representative values in http://www.spec.org/cpu2000/results/cfp2000.html ? IIRC, Platform says they derive their numbers from spec anyway. besides, these numbers make very little difference - don't they simply bias the scheduler towards running jobs on faster machines first? in my experience, users will have more jobs queued than you can run, so every machine will remain busy and the speed-bias doesn't matter.
- Previous message: [Beowulf] LSF and cpu factor
- Next message: [Beowulf] lsf mpi job run policy !!!
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
More information about the Beowulf mailing list
