Adaptive Passes are terminating abruptly in 2021 R1 when solved in Distributed Fashion
-
-
I am trying to simulate an multiport network in HFSS. As single machine alone wouldn't fit memory requirements, I opted to use distributed computing feature. So I setup 3 machines for simulation using Intel MPI. But the adaptive meshing process got terminated at second pass without stating the detailed reason for simulation error [tried multiple times and did reinstallation couple of times in all machines, error ceased to disappear :( ]. Then I reduced the simulation volume to see if issue persists, and it actually does. Somehow I managed to finish the adaptive passes on a high memory machine and thought of giving a shot to see if the error occurs in solving sweep frequencies too. This time it got distributed without any issues and solved all sweep frequencies as expected. I wanted to see if the issue is more related to MPI vendor, as in previous version (2020 R1) was using the platform-computing MPI to perform simulations. Simulation got completed without any errors in 2020 R1 with same Intel MPI distribution. I am baffled why the 2021 R1 is stuck with the adaptive passes, I would appreciate your help if you let me know how to get rid of this.
P.S. : I tried both auto and manual HPC settings and verified all the MPI test commands. Firewalls are completely turned off. :)
Please see the error attached in figure.
Viewing 2 reply threads- The topic ‘Adaptive Passes are terminating abruptly in 2021 R1 when solved in Distributed Fashion’ is closed to new replies.
-
You are navigating away from the AIS Discovery experience