[GE users] One slot per host, or a way to consume all slots

Reuti reuti at staff.uni-marburg.de
Sat Feb 24 10:42:25 GMT 2007


Florian:

Am 23.02.2007 um 18:40 schrieb FL:

> I see from the link that I need to say more about gaussian parallel  
> jobs.

we are also using Gaussian03 :-)

> This method will work when the user submits a single-threaded job.

You mean you want the complete machine for the job because of memory  
constraints?

> However, I do have a parallel execution environment for gaussian  
> parallel
> jobs, and $pe_slots is set to the number of workerjobs-1; I see  
> that parallel

To avoid misunderstandings: with $pe_slots I meant the setting in the  
definition of the PE. I.e. all slots must come from one and only one  
machine.

> gaussian jobs would need to set $pe_slots to (max slots per host)* 
> (workerjobs-1)
> in order to consume the available slots in that case..

This formula I don't get: why are you using (workerjobs-1)? If you  
refer to Lindaworkers, a parallel Gaussian job with a proper % 
lindaworker list and 4 cpu requested would also need "-pe linda 4".

To lay out your intention, you mean with a cluster of:

node1: 4 cores, 16 GB RAM
node2: 4 cores, 16 GB RAM
node3: 2 cores, 8 GB RAM
node4: 2 cores, 8 GB RAM

to start just one Gaussian job with 4 processes, hence with one  
process running per node only?

-- Reuti

PS: Just to note, that there are some G03 tips on the list: http:// 
gridengine.sunsource.net/servlets/ReadMsg?listName=users&msgNo=14600


> .this makes the job submission
> script explicitly dependent on the number of slots per host, unless  
> I misunderstand...
>
> On 2/23/07, Reuti <reuti at staff.uni-marburg.de> wrote: http:// 
> gridengine.sunsource.net/servlets/ReadMsg?list=users&msgNo=18988
>
> -- Reuti
>
>
> Am 23.02.2007 um 17:26 schrieb FL:
>
> > I'm running SGE 6.0u7 on a linux cluster; the OS is  RHEL 4.
> >
> > We have Gaussian G03 computational chemistry jobs that
> > should have the following property with respect to SGE:
> > if a G03 job is running on a node, then it should be the ONLY
> > job running on that node.  It's not enough for there to be
> > other (unrelated) suspended jobs on the node: making every
> > "non-Gaussian" queue subordinate to a "Gaussian" queue
> > won't be satisfactory (we've tried this and it tends to starve
> > the G03 jobs).
> >
> > I know that one way to ensure that G03 is the only job running
> > on a node if it is running is to set the number of slots per host
> > to one
> > (I could use a reference for this). However, that limits the
> > usefulness
> > of the node for jobs that can coexist with other jobs.
> >
> > is there a way to ensure that if a job in a certain queue
> > runs, then it consumes all the available slots? Or is there a better
> > way to ensure that if a job of a certain type is running on a node,
> > it is the only job running there?
> >
> > Thanks,
> > FL
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: users-unsubscribe at gridengine.sunsource.net
> For additional commands, e-mail: users-help at gridengine.sunsource.net
>
>

---------------------------------------------------------------------
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