On May 2, 2013, at 23:45 , Reuti wrote:

> Hi,
> 
> Am 02.05.2013 um 15:47 schrieb Esztermann, Ansgar:
> 
>> can someone tell me what units are used by sge_share_mon (or the share tree 
>> part of qmon)? I had assumed CPU-seconds, Gigabytes-seconds and Gigabytes 
>> (as usual), but (at least) the CPU values are often much too large -- I've 
>> just run qconf -clearusage, and two hours later see a thousand CPU-years for 
>> one user who is occupying only about 1000 slots.
>> 
>> Am I overlooking something?
> 
> To which column do you refer - cpu? In which range is the outpout - can you 
> please reset the usage and run a fixed number of endless loops or alike for a 
> set h_rt and reproduce the output?

Curiouser and curiouser:
Since the reset of eight days ago, the 12k core cluster has again accumulated 
more than 60k CPU-years my own tests (4 cores for 5 minutes and 48 cores for 5 
hours) yield exactly the expected results, though.
I will test multi-node jobs next.


A.

-- 
Ansgar Esztermann
DV-Systemadministration
Max-Planck-Institut für biophysikalische Chemie, Abteilung 105

Attachment: smime.p7s
Description: S/MIME cryptographic signature

_______________________________________________
users mailing list
[email protected]
https://gridengine.org/mailman/listinfo/users

Reply via email to