[torqueusers] what version of torque to upgrade to?

Michel Béland michel.beland at calculquebec.ca
Tue Feb 12 08:30:06 MST 2013


John Valdes wrote:

> All,
>
> We've been using Torque 2.3.x and Maui 3.2.6px on our modest size
> (~350 nodes), production, commodity cluster successfully now for the
> last 3 years or so, and while we have encountered minor bugs every now
> and then, for the most part it has been very stable and reliable.
> Nevertheless, we're thinking that we should upgrade to a current
> version of Torque and Maui (3.3.1), partly so that we're using an
> activately maintained codebase, but also to get cgroup and better GPU
> support.

If you intend to use the -lnodes=1:gpus=2 syntax with Maui, you have to 
be aware that Maui does not understand this syntax. It cannot schedule 
GPU jobs like these.

On a small GPU cluster that we have here, we ended up asking users to 
use -lnodes=1:ppn=2 to request gpus, as we have one gpu per core on our 
nodes. This has some shortcomings but we decided that we could live with 
them.

We also tried generic resources, but it failed miserably. There are 
patches floating around to manage gpus with Maui, but I could never make 
them work better than my imperfect solution.

-- 
Michel Béland, analyste en calcul scientifique
michel.beland at calculquebec.ca
bureau S-250, pavillon Roger-Gaudry (principal), Université de Montréal
téléphone : 514 343-6111 poste 3892     télécopieur : 514 343-2155
Calcul Québec (www.calculquebec.ca)
Calcul Canada (calculcanada.ca)



More information about the torqueusers mailing list