Multicore post-processing of solutions (from another study) in parallel preferably without GUI

Please login with a confirmed email address before reporting spam

The below post is related to an archived discussion


I am trying to do post processing of solutions coming from an eigenvalue study that has already generated an enormous file. I am using the Derived Values interface under Results to do volume integration of modes. Unfortunately, I am only able to do it with the GUI and it is using only one core (out of 64) and one solution at a time, which is very slow.

Is there a way to do this in parallel and multicore, ideally without the use of GUI, such that a txt file is generated after the process?

My version of COMSOL is 5.4.


2 Replies Last Post 15 oct. 2024, 02:31 UTC−4
Edgar J. Kaiser Certified Consultant

Please login with a confirmed email address before reporting spam

Posted: 2 months ago 14 oct. 2024, 14:54 UTC−4

Hi Ali,

post processing can be a little painful in between. I sometimes used a workaround for integrations by doing them during computations in nonlocal couplings. It is then important to store the result in a state variable which becomes part of the solution. If you don't use a state variable COMSOL will redo the integration in the post processing. In my experience the solution is not significantly slowed down by this procedure. Solver code is more efficient than post processing code.

Other people were exporting the solution in order to do such post processing in own software, e.g. Python. Not sure if this is faster though. I never tried it.

Cheers Edgar

-------------------
Edgar J. Kaiser
emPhys Physical Technology
www.emphys.com
Hi Ali, post processing can be a little painful in between. I sometimes used a workaround for integrations by doing them during computations in nonlocal couplings. It is then important to store the result in a state variable which becomes part of the solution. If you don't use a state variable COMSOL will redo the integration in the post processing. In my experience the solution is not significantly slowed down by this procedure. Solver code is more efficient than post processing code. Other people were exporting the solution in order to do such post processing in own software, e.g. Python. Not sure if this is faster though. I never tried it. Cheers Edgar

Gunnar Andersson COMSOL Employee

Please login with a confirmed email address before reporting spam

Posted: 2 months ago 15 oct. 2024, 02:31 UTC−4

Computing volume integrals in Results has been parallelized for a couple of versions, but I don't think that it was parallelized in 5.4.

Computing volume integrals in Results has been parallelized for a couple of versions, but I don't think that it was parallelized in 5.4.

Reply

Please read the discussion forum rules before posting.

Please log in to post a reply.

Note that while COMSOL employees may participate in the discussion forum, COMSOL® software users who are on-subscription should submit their questions via the Support Center for a more comprehensive response from the Technical Support team.