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

[Linux-cluster] Running Parallel Jobs Without MPI



Hi everyone,
I am more or less newbie with clusters so forgive me if the question
is retarded.
I was wondering if it is at all possible  to run *parallel* jobs with
code that DOES NOT  use MPI libraries. I often reuse old 'homemade'
code or use someone elses code I found online. Most of the time this
is a good code but too complex and incorporating MPI would take me
forever. I would  like to be able to run it in parallel without
modification, even if it does run somewhat slower than it would with
MPIs.  Is there a way to do this?

I know PBS/Torque has command for this (pbdsh), but installing PBS on
a small 12 nodes cluster only to to run parallel jobs seems like an
overkill.

I also saw the pdsh tool, but I wasn't entirelly sure if this is just
a sysadmin tool.
(So that administrators, for example, can just go
mkdir foo
on all nodes at once rather then doing it 50 times.) Or is this
exactly what I am looking for?

Thanks for your help


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