[torquedev] [Bug 78] fix routing of jobs based on nodes spec

bugzilla-daemon at supercluster.org bugzilla-daemon at supercluster.org
Tue Aug 31 16:23:17 MDT 2010


Ken Nielson <knielson at adaptivecomputing.com> changed:

           What    |Removed                     |Added
             Status|NEW                         |RESOLVED
                 CC|                            |knielson at adaptivecomputing.
                   |                            |com
         Resolution|                            |WONTFIX

--- Comment #4 from Ken Nielson <knielson at adaptivecomputing.com> 2010-08-31 16:23:16 MDT ---

We are not going to accept this patch for a couple of reasons.

The first reason is we are planning to address the syntax issue soon. Instead
of adding an additional syntax that will likely be overridden shortly we will
be better off addressing the global issue.

The second reason is that allowing a nodes=x:ppn=y on the resources_max.nodes
and resources_min.nodes creates and ambiguity. The solution you have proposed
changes the meaning of nodes to processors. While schedulers like Moab will
allow nodes and processors to mean the same thing, Moab handles the
interpretation. TORQUE still looks at everything as nodes first and processors
as a resource on a node.

It might be useful to have a way to set queue limits with the nodes=x:ppn=y
concept. But we will address that later.

Thanks for the patch. The help from the community is vital to the success of
the TORQUE project.

Configure bugmail: http://www.clusterresources.com/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

More information about the torquedev mailing list