minor bug! Re: [torqueusers] TORQUE 2.0 Release

Ole Holm Nielsen Ole.H.Nielsen at fysik.dtu.dk
Wed Oct 26 02:22:32 MDT 2005


My 5 cents worth: I've followed the official advice in
http://www.clusterresources.com/products/torque/docs20/a.flargeclusters.shtml
and compiled Torque with --disable-rpp.  Our cluster has 941
nodes and I haven't noticed any communications problems.
We use Torque 1.2.0p6.

Regards,
Ole

Roy Dragseth wrote:
>>If you *don't* configure with --disable-rpp, quick startup is broken
>>without this:
> 
> 
> Have anyone any real experience to share over what scales best of rpp or tcp?  
> I've seen comments going both ways and is a bit confused.  I need to make a 
> decision on what to make the default in the pbs setup for Rocks, some of the 
> users of the pbs setup for rocks is moving towards 1000node clusters and I 
> need to make sure they get the best solution.

-- 
Ole Holm Nielsen
Department of Physics, Technical University of Denmark


More information about the torqueusers mailing list