Hi,
Although I have run some mpi jobs (NAS benchmark), the condor_userprio shows this information which are vague for me
# condor_userprio -allusers
Last Priority Update: 2/9 14:22
Effective Priority Res Total Usage Time Since
User Name Priority Factor In Use (wghted-hrs) Last Usage
------------------------------------------- ------------ --------- ------ ------------ ----------
DedicatedScheduler@xxxxxxxxxxxxxxxxxxxxxxxx 500.00 1000.00 0 0.93 0+01:40
mahmood@local 664.54 1000.00 4 14.49 <now>
------------------------------------------- ------------ --------- ------ ------------ ----------
Number of users: 2 4 15.42
Although it says two users, it is actually one user (me) which submits the jobs. Ignoring the current job, total previous jobs took about 4400 seconds.
bt.C.4 => 703.06 sec
ep.D.4 => 2509.88 sec
cg.C.2 => 177.56 sec
ft.C.2 => 289.30 sec
ft.C.4 => 157.86 sec
lu.C.4 => 557.21 sec
The 4 and 2 are the number of cores (machine count) for that job. So, I expect that
4*703 + 4*2509 + 2*177 + 2*289 + 4*157 + 4*557 = 16636 sec = 4.6 hours
That means, the user (mahmood) has used 4.6 hours of the system running those jobs regarding the number of cores engaged. But I don't see that in the stats.
Can someone shed a light on that?
Regards,
Mahmood