[GE users] job_load_threshold under 6.0u1

Stephan Grell - Sun Germany - SSG - Software Engineer stephan.grell at sun.com
Thu Nov 4 08:35:30 GMT 2004


Hi,

sorry, for replying sooooo late.

There is a misunderstanding in how the load adjustments work. It is
a scheduler internal feature and has nothing to do with the rest of the
system (that is why the configuration parameters are part of the scheduler
configuration ;-) ).
The job_load_adjustments mean, that when a job is dispatched to a host,
it will add that amount of load to that host (in your example: np_load_avg
would increase by 1.0). Since it will take some time, before the job is
started
and the load sensor reports the new load, one can simulate that load with
the help of the load adjustments.
The decay time is the time interval, after which the simulated load is
gone and
one is sure, that the job is started and the real load is correct.

The purpose of this feature is to prevent an overload situation on the
hosts and
to ensure, that a queue will not go into alarm state based on the jobs
dispatched
by the scheduler.

To ensure, that the load_adjustments are working correct, take your
example and
assign that queue 10 slots. Submit 10 jobs into the system and you will
see, that
only one will be started.

Does this help?

Cheers,
Stephan

Sean Dilda wrote:

>On Thu, 2004-10-07 at 09:30, Reuti wrote:
>  
>
>>Hi, 
>> 
>>    
>>
>>>job_load_adjustments              np_load_avg=1.0 
>>>      
>>>
>> 
>>np_load_avg is calculated, and maybe therefore the adjustment is not working. 
>>Did you tried to adjust the load_avg instead also? 
>>    
>>
>
>Thanks for the suggestion.  I tried it with load_avg instead and got the
>same thing.  I was unable to make the queue hit alarm state, and qhost
>-F never showed load_avg or np_load_avg with any value other than 0.0
>
>
>---------------------------------------------------------------------
>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