This may be more what you want to do: http://abaddogday.blogspot.com/2011/03/how-to-configure-gridengine-with.html
I've set this up for many customers with your issue. You want to prevent oversubscribing your nodes memory... -Bill On Mon, Feb 23, 2015 at 4:57 PM, Ed Lauzier <[email protected]> wrote: > Hi, > > Increase the default h_vmem to 4G or 6G. > vmem is different that res mem, and it is the only practical memory > limit that can be enforced by current linux kernels with opensource grid > engine. > > For java jvms, you may need to set your mx value to the max vmem value you > use for default. > > For normal running programs, vmem is slightly greater than res mem. With > other programs > or packages, h_vmem can be 2x or 3x that of res mem, or more if programs > do mmap. > > You can easily test this with qlogin -l h_vmem=4G and then test your app > directly. > > For more fine grandular control of your runtime resource limits, look into > Univa > Grid Engine specifically their cgroups integration... > > The easiest way to deal with the issue is to purchase machines with a lot > of memory... :) > if you can.... > > -Ed > > > -----Original Message----- > *From:* Mishkin Derakhshan [mailto:[email protected]] > *Sent:* Monday, February 23, 2015 07:07 PM > *To:* [email protected] > *Subject:* [gridengine users] How to set up h_vmem as a consumable > resource > > Hi, > We have some jobs that require significant amounts of memory so we want to > try and setup h_vmem as a consumable resource to manage this. > > This is what we have setup: > $ qconf -sq dev.q | grep h_vmem > h_vmem 3.7G > > $ qconf -sc | grep h_vmem > h_vmem h_vmem MEMORY <= YES YES > 0 0 > > And if we submit jobs like this then we don't have any problems, > $ qsub -b y -j y -l h_vmem=1G -q dev.q sleep 100 > > But if we submit jobs without explicitly requesting h_vmem (i.e., we > don't use -l h_vmem=X) then the jobs die on startup saying it can't > allocate memory: > error reason 1: 02/19/2015 14:13:39 [0:14840]: can't set > additional group id (uid=0, euid=0): Cannot allocate memory > > We _think_ this has to do with setting a default h_vmem (on a queue basis? > host basis?) so jobs that don't explicitly request the resource will use > something by default, but we've been unable to figure out how to set this > up. > > We are using 6.1u3. > > thanks > > > _______________________________________________ > users mailing list > [email protected] > https://gridengine.org/mailman/listinfo/users > >
_______________________________________________ users mailing list [email protected] https://gridengine.org/mailman/listinfo/users
