thats strange.
please provide more information on what sort of jobs did u submit both times?
were the jobs identical or different?
also specify if the execute machines were the same in both the experiments.
Chris Miles <chrismiles@xxxxxxxxxxxxxxxx> wrote:
I ran an experiment today.
50 jobs submitted to a 20 machine condor pool over GCB.
Job completion times were roughly the same (10 seconds deviati!
on) when
submitting to
the pool with 10 machines (20vms) then when submitting to 20 machines (40vms).
Surely this cant be normal behaivour?
Chris
University of Paisley, Scotland
_______________________________________________
Condor-users mailing list
Condor-users@xxxxxxxxxxx
https://lists.cs.wisc.edu/mailman/listinfo/condor-users
![[Computer Systems Lab]](http://www.cs.wisc.edu/pics/csl_logo.gif){kind=logo}