> > It would be more straight forward to use directly h_vmem. This is controlled > by SGE and the job exceeding the limit will be killed by SGE. If you consume > it as a consumable on a exechost level, it could be set to the built in > physical memory. >
That's what I use - h_vmem is a consumable resource, and a JSV that rejects jobs that do not include an h_vmem "request". If the jobs run over there limit, SGE takes care of it. I also cheat a little, and if they forget to request memory, the config automatically drops a default 4G request in. This allows me to ensure that on a 12 core box and 48GB of RAM, I can run nearly 12 jobs keeping the desired 1 per core. Ian -- Ian Kaufman Research Systems Administrator UC San Diego, Jacobs School of Engineering ikaufman AT ucsd DOT edu _______________________________________________ users mailing list [email protected] https://gridengine.org/mailman/listinfo/users
