Re: [HTCondor-users] Parallel Universe Job Start

[Gabriel Mateescu <mateescu@xxxxxxx>]

I believe that, once all the resources required
by a parallel job have been claimed, the time
needed to start the tasks of the job is no longer
determined by the negotiator cycle, but by the
delay of communication between the scheduler
daemon and all the worker nodes belonging to
the job.

If communication is serial, then this takes time
O(p), where p is the number of worker nodes for
the job.

Gabriel




On Sat, Jan 17, 2015 at 6:38 AM, Philip Papadopoulos
<philip.papadopoulos@xxxxxxxxx> wrote:
> Suppose I have a parallel universe job.
> Now suppose all the parallel resources needed by the job are actively
> claimed.
>
> What is the longest time delay between the -start- of execution of the first
> task
> and the -start- of execution of the last task?
>
> My intuition says it is either one or two cycles of the negotiator O(10
> min), but I'm probably completely off base
>
> Thanks,
> Phil
>
>
> --
> Philip Papadopoulos, PhD
> University of California, San Diego
> 858-822-3628 (Ofc)
> 619-331-2990 (Fax)
>
> _______________________________________________
> 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/



-- 
Gabriel Mateescu
Email: mateescu@xxxxxxx
Web: https://www.linkedin.com/in/mateescu