Long-term simulation and RAM demand

[Juan Faundez]

Dear all

I'm trying to run a simulation for 30 years, however, the simulation is killed after just 10 years. I'm demanding from the HPC 5 nodes (44 cores each one) and 17400 Ram-per-mem, and it is not enough. I tried canceling the traj.nc file (false), but not yet with a decrease in the number of particles (30.000 twice per month). Our HPC support is asking where we could manage the limit of RAM demanded within the CMS code because it seems unlimited. Or someone could recommend me a configuration to avoid the blow-up

thank you by your advice

best