[GE users] changing parameters while running...

John Tseng jtseng at montalvosystems.com
Mon Aug 28 17:03:48 BST 2006


    [ The following text is in the "utf-8" character set. ]
    [ Your display is set for the "ISO-8859-10" character set.  ]
    [ Some characters may be displayed incorrectly. ]

There are two max job per user setting:

max_u_jobs:  maximum number submitted - this is a sge qmaster restriction
maxujobs: maximum number of user jobs running. this is a scheduler restriction

Changing maxujobs (sched) has no affect on running jobs.  The scheduler will simply notice the threshold is exceeded and not schedule any more jobs.
Changing max_u_jobs (queue) could prevent people from submitting more jobs.

Usually we don't limit max_u_jobs, but we change maxujobs as approrpriate.

One reason to change max_u_jobs (queue) is if you can't handle thousands of job in a scheduling cycle due to memory or cpu restrictions.
At that point, it would be best to move to a job task array to simplify scheduling, or have users self throttle their submissions.

-john

details...

man sge_conf:
max_u_jobs
       The  number  of  active (not finished) jobs which each Grid Engine user
       can have in the system simultaneously is controlled by this  parameter.
       A  value  greater  than  0 defines the limit. The default value 0 means
       "unlimited". If the max_u_jobs limit is exceeded by  a  job  submission
       then  the submission command exits with exit status 25 and an appropri-
       ate error message.

       Changing max_u_jobs will take immediate effect.

       This value is a global configuration parameter only. It cannot be over-
       written by the execution host local configuration.


man sched_conf
  maxujobs
       The  maximum  number of jobs any user may have running in a Grid Engine
       cluster at the same time. If set to 0 (default) the users  may  run  an
       arbitrary number of jobs


On Monday 28 August 2006 06:05, Davide Cittaro wrote:
> Hi there, I see there are people running their jobs on the little  
> cluster... I have to change and reduce a value (max jobs/users) but I  
> don't know what can happen now:
> [a] jobs running continue until their end (-> I can safely set it now)
> [b] some jobs will be stopped to satisfy the new parameter... (-> I  
> have to wait the jobs to end)
> 
> And yes, the number of jobs/user is now higher that the one I'm going  
> to set set...
> 
> thanks
> 
> 
> /*
> Davide Cittaro
> HPC and Bioinformatics Systems @ Informatics Core
> 
> IFOM - Istituto FIRC di Oncologia Molecolare
> via adamello, 16
> 20139 Milano
> Italy
> 
> tel.: +39(02)574303355
> e-mail: davide.cittaro at ifom-ieo-campus.it
> */
> 
> 
> 
> 

---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe at gridengine.sunsource.net
For additional commands, e-mail: users-help at gridengine.sunsource.net




More information about the gridengine-users mailing list