|November 13, 2009, 04:34||
How to start parallelized computation in multi-nodes?
Join Date: Mar 2009
Posts: 116Rep Power: 9
I hope all of you are fine.
Now I am also trying to compute flow field in multi-nodes.
But there are some block on my way.
For openfoam, in the user guide, I need to execute mpirun in multi-nodes as follows:
mpirun --hostfile <machines> -np <nProcs>
<foamExec> <otherArgs> -parallel > log
and the machines file should include the name of nodes.
But the cluster in my institute is that I need to write the script as follows:
#PBS -l nodes=2pn=4,walltime=01:02:00.00 -q helics2
mpiexec -verbose program_to_run [arguments] [< input] [> output] So, it is totally different. Do you ever encounter this problem? How do you deal with it?
Does openfoam have its own way for parallelized computing or it is the same as OpenMPI?
Any suggestion from you is useful for me and I will be grateful for that.
By the way, I am still working on the that problem.
|Thread||Thread Starter||Forum||Replies||Last Post|
|TimeVaryingMappedFixedValue||irishdave||OpenFOAM Running, Solving & CFD||28||May 28, 2015 13:37|
|Interpolation (Number of mapped nodes)||Sam||CFX||3||January 6, 2008 20:51|
|meshing F1 front wing||Steve||FLUENT||0||April 17, 2003 12:37|
|CFX4.3 -build analysis form||Chie Min||CFX||5||July 12, 2001 23:19|
|Lets start the public domain CFD-Project!||Heinz Wilkening||Main CFD Forum||3||March 11, 1999 23:55|