Parallel Computing decomposePar

Persia · November 4, 2016, 13:59

-n specifies the number of nodes.
-np specifies the number of processes.

I cannot comment on your inquiry but I guess it should not matter which one (-n or -np) you use as long as you specify the same number of processes/nodes.

Prakriti · March 3, 2020, 00:39

Hi all,

I've been using blockMeshDict and thereafter snappyHexMeshDict for a given case to run in parallel.

In the blockMeshDict, when my base hex resolution is (40 40 40), mesh decomposition works fine. However, when I increase this base resolution to (128 128 128), and run the case, blockMesh and snappyHexMesh create the mesh, decomposePar runs without returning any error messages, and the simulation starts. But when I run "reconstructParMesh", it gives me a "no mesh" error at EACH timestep.

Has anyone encountered this odd error and if so, how is it to be navigated?

marco.müller · March 3, 2020, 01:27

Hi Praktiki,

there are many possible error sources in such a long process chain.

Is there a specific reason why you are using reconstructParMesh? I've stopped using it because I also encountered problems with it. Also also remember that it is still some beta version?!

For post processing in Paraview as far as I know you dont need it.

Marco

March 3, 2020, 00:39	reconstructParMesh returns "no mesh" message.	#42
Prakriti New Member Prakriti Join Date: Jun 2018 Posts: 6 Rep Power: 7	Hi all, I've been using blockMeshDict and thereafter snappyHexMeshDict for a given case to run in parallel. In the blockMeshDict, when my base hex resolution is (40 40 40), mesh decomposition works fine. However, when I increase this base resolution to (128 128 128), and run the case, blockMesh and snappyHexMesh create the mesh, decomposePar runs without returning any error messages, and the simulation starts. But when I run "reconstructParMesh", it gives me a "no mesh" error at EACH timestep. Has anyone encountered this odd error and if so, how is it to be navigated?

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
unchangeable continuity residuals in parallel computing	wlt_1985	FLUENT	0	August 1, 2011 12:15
Diffusion equation solved using Parallel Computing	Sachin Paramane	Main CFD Forum	0	June 11, 2007 23:48
Parallel Computing on Multi-Core Processors	Upgrading Hardware	CFX	6	June 7, 2007 15:54
Parallel Computing	peter	Main CFD Forum	7	May 15, 2006 09:53
Parallel Computing Classes at San Diego Supercomputer Center Jan. 20-22	Amitava Majumdar	Main CFD Forum	0	January 5, 1999 12:00

November 4, 2016, 13:59		#41
Persia New Member Maryam Join Date: Dec 2015 Posts: 13 Rep Power: 10	-n specifies the number of nodes. -np specifies the number of processes. I cannot comment on your inquiry but I guess it should not matter which one (-n or -np) you use as long as you specify the same number of processes/nodes.

March 3, 2020, 01:27		#43
marco.müller New Member Marco Müller Join Date: Feb 2013 Location: Germany Posts: 23 Rep Power: 13	Hi Praktiki, there are many possible error sources in such a long process chain. Is there a specific reason why you are using reconstructParMesh? I've stopped using it because I also encountered problems with it. Also also remember that it is still some beta version?! For post processing in Paraview as far as I know you dont need it. Marco