Daniele Varsano wrote: ↑Thu May 26, 2022 9:41 am
2) The input I posted is just an example for a run with 16 mpi, you need to adapt to the resources of your run. The product of the CPU assigned for each role needs to match the number of MPI tasks. Note that this is general for yambo and not related at all to the new cutoff and rimW scheme.
e,g for 32 CPU:
Code: Select all
X_and_IO_CPU= "1 1 1 16 2" # [PARALLEL] CPUs for each role
X_and_IO_ROLEs= "q g k c v" # [PARALLEL] CPUs roles (q,g,k,c,v)
DIP_CPU= "1 16 2" # [PARALLEL] CPUs for each role
DIP_ROLEs= "k c v" # [PARALLEL] CPUs roles (k,c,v)
SE_CPU= "1 1 32" # [PARALLEL] CPUs for each role
SE_ROLEs= "q qp b" # [PARALLEL] CPUs roles (q,qp,b)
Here you can find a tutorial on yambo parallel strategies:
http://www.yambo-code.org/wiki/index.ph ... strategies
Best,
Daniele
Dear Daniele sir,
I am struggling with the parallel running of GW-BSE calculation. I prepared a GW input file
G0W0file.txt
. Then I will do the BSE part.
Can You please confirm my input for parallel execution in G0W0 input file;
X_Threads= 0 # [OPENMP/X] Number of threads for response functions
X_CPU= "1 1 1 16 2" # [PARALLEL] CPUs for each rolei
X_ROLEs= "q g k c v" # [PARALLEL] CPUs roles (q,g,k,c,v)
#X_nCPU_LinAlg_INV= $ncpu # [PARALLEL] CPUs for Linear Algebra
X_Threads= 0 # [OPENMP/X] Number of threads for response functions
DIP_Threads= 0 # [OPENMP/X] Number of threads for dipoles
SE_CPU= "1 1 32" # [PARALLEL] CPUs for each role
SE_ROLEs= "q qp b" # [PARALLEL] CPUs roles (q,qp,b)
SE_Threads= 0
I have tried to use the suggested method in wiki to write a number of cores for running at 32 total CPUs. I would like to run this using the attached slurm script
slurm_script.txt
in our HPC.
Is this parallel description also need to be included with slurm script or there can be just mpi command without these tags?
Thank You
Regards
Sumit Kukreti
PhD scholar
IIT Jodhpur INDIA