[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [HTCondor-users] Whole memory request for wholeMemory job.



Hi Salman,

A more up-to-date version of the wiki page you landed on is here:

https://htcondor-wiki.cs.wisc.edu/index.cgi/wiki?p=WholeMachineSlots

As noted on that page, you may want to look into the new way to support a mix of single-core and full machine jobs, which is partitionable slots.

--Dan

On 6/3/13 4:26 AM, Salman Toor wrote:
Hi, 

I am new to Condor but still mange to run the basic setup without any problem. 

Now I want to allow my users to submit full machine jobs. I want to use the same setup, I mean if the whole-machine job comes to a resource and other single slot jobs are running condor simply suspend the wholemachine job and start it once the single machine jobs finished. With the help of following document I have managed to to this 


Now things are working fine, but it just turned out that when a full machine job get started on slot1 (rest of the slots are empty) still it uses the memory assigned only to slot1... 
I mean I accept that wholemachine job should have wholememory as well. 

Can someone please guide me how do I fix this? Following is my configuration. 

----------------
#require that whole-machine jobs only match to Slot1
START = ($(START)) && (TARGET.RequiresWholeMachine =!= TRUE || SlotID == 1)

# have the machine advertise when it is running a whole-machine job
STARTD_JOB_EXPRS = $(STARTD_JOB_EXPRS) RequiresWholeMachine

# Export the job expr to all other slots
STARTD_SLOT_EXPRS = RequiresWholeMachine

# Suspend the whole-machine job until the other slots are empty
SUSPEND = ($(SUSPEND)) || (SlotID == 1 && Slot1_RequiresWholeMachine =?= True && \
           (Slot2_Activity =?= "Busy" || Slot3_Activity =?= "Busy" || Slot4_Activity =?= "Busy" ) )

# suspend existing single-cpu jobs when there is a whole-machine job
SUSPEND = ($(SUSPEND)) || (SlotID != 1 && Slot1_RequiresWholeMachine =?= True)

CONTINUE = ( $(SUSPEND) =!= True )
---------------

Regards..
Salman. 

 

Salman Toor





_______________________________________________
HTCondor-users mailing list
To unsubscribe, send a message to htcondor-users-request@xxxxxxxxxxx with a
subject: Unsubscribe
You can also unsubscribe by visiting
https://lists.cs.wisc.edu/mailman/listinfo/htcondor-users

The archives can be found at:
https://lists.cs.wisc.edu/archive/htcondor-users/