[Condor-users] Scheduler for parallel universe

["Steven Platt" <Steven.Platt@xxxxxxxxxx>]

Apologies if this gets multiple postings ... this is the 3rd attempt to
send this message!

Hi,

Having looked through sections 2.9 & 3.12.8 of the Condor manual I
understand that I need to set up a dedicated scheduler to enable
parallel universe & MPI jobs. I want to allow both dedicated and
opportunistic jobs on the same set of machines.

Questions:
-- Do I need to setup/specify a new scheduler, or can I use the one
already in place on the central manager?
-- If a new scheduler is necessary, it needs to be on a separate box
from the current scheduler doesn't it?

My small cluster:
Condor 7.0.5 on Rocks 5.1
48 slots on 8 CentOS 5.2 boxes (execute nodes only); central manager is
CentOS 5.2 that's also the only submit node. Very low job throughput and
no interactive users so pre-emption, suspension & rank clashes are not
an issue.


Thanks

Steve

Dr Steven Platt
Statistics, Modelling and Bioinformatics
Health Protection Agency
London
-----------------------------------------
**************************************************************************
The information contained in the EMail and any attachments is
confidential and intended solely and for the attention and use of
the named addressee(s). It may not be disclosed to any other person
without the express authority of the HPA, or the intended
recipient, or both. If you are not the intended recipient, you must
not disclose, copy, distribute or retain this message or any part
of it. This footnote also confirms that this EMail has been swept
for computer viruses, but please re-sweep any attachments before
opening or saving. HTTP://www.HPA.org.uk
**************************************************************************