Note: This discussion is about an older version of the COMSOL Multiphysics® software. The information provided may be out of date.
Discussion Closed This discussion was created more than 6 months ago and has been closed. To start a new discussion with a link back to this one, click here.
Problem with solving large models
Posted 06.03.2013, 21:00 GMT-5 Studies & Solvers Version 4.2a 1 Reply
Please login with a confirmed email address before reporting spam
Hi,
Lately, I've been trying to solve transient, non-linear heat conduction problems with ~500,000 elements. I've noticed that my computer seems to stall during the solution step and was interested to know what might be happening and how to resolve it. I'm using Redhat Linux. When I use the command "top," I can see that 80+% of my memory is being used but <10% of my CPUs are being used.
Do you know why Comsol is having a problem?
Thanks,
Kevin
Lately, I've been trying to solve transient, non-linear heat conduction problems with ~500,000 elements. I've noticed that my computer seems to stall during the solution step and was interested to know what might be happening and how to resolve it. I'm using Redhat Linux. When I use the command "top," I can see that 80+% of my memory is being used but <10% of my CPUs are being used.
Do you know why Comsol is having a problem?
Thanks,
Kevin
1 Reply Last Post 07.03.2013, 00:58 GMT-5
Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
Hi
well it depends if the PC is shuffling to disk or not, that makes a big difference, 0.5MDoF can take some time, and depending in which phase of the solving process COMSOL is it uses 1 to N processors. One way is to reduce the number of elements and run again, just to be sure the results are more or less OK and the solver gets to it's end, then remesh finer and wait ...
--
Good luck
Ivar
well it depends if the PC is shuffling to disk or not, that makes a big difference, 0.5MDoF can take some time, and depending in which phase of the solving process COMSOL is it uses 1 to N processors. One way is to reduce the number of elements and run again, just to be sure the results are more or less OK and the solver gets to it's end, then remesh finer and wait ...
--
Good luck
Ivar