Optimization of HPC job

raunakjung · June 1, 2017, 18:07

I am using I am using a HPC
Apro green server
Centos 6.3 Linux based
256 core

Since the HPC has time limit 72:00:00 I am not able to finish my calculation within the time. How could I optimize my resources use to finish job in time.
I can use Distributed MPI parallel for my calculations

Thank You

Following is my script file :
#@ resources = ConsumableCpus (1) ConsumableMemory(2gb)

#@ wall_clock_limit =72:00:00

#@ total_tasks =32

#@ blocking =unlimited
. /applic/Applications/env_setup/cfx_v150.env

nodes=`cat $LOADL_HOSTFILE`
nodes=`echo $nodes | sed -e 's/ /,/g'`

export CFX5RSH=ssh

cfx5solve -mdef CFX.mdef -start-method "Platform MPI Distributed Parallel" -par -def

evcelica · June 5, 2017, 13:20

You can set wall clock limit in CFX Pre. So the simulation will stop at say...71 hours.
Can you then just restart the simulation using the previous results file as initial conditions? Therefore, just breaking it into two+ runs?

June 1, 2017, 18:07	Optimization of HPC job	#1
raunakjung Senior Member raunak jung pandey Join Date: Jun 2016 Posts: 102 Rep Power: 9	I am using I am using a HPC Apro green server Centos 6.3 Linux based 256 core Since the HPC has time limit 72:00:00 I am not able to finish my calculation within the time. How could I optimize my resources use to finish job in time. I can use Distributed MPI parallel for my calculations Thank You Following is my script file : #@ resources = ConsumableCpus (1) ConsumableMemory(2gb) #@ wall_clock_limit =72:00:00 #@ total_tasks =32 #@ blocking =unlimited . /applic/Applications/env_setup/cfx_v150.env nodes=`cat $LOADL_HOSTFILE` nodes=`echo $nodes \| sed -e 's/ /,/g'` export CFX5RSH=ssh cfx5solve -mdef CFX.mdef -start-method "Platform MPI Distributed Parallel" -par -def

Thread Tools	Search this Thread
Show Printable Version Email this Page	Search this Thread: Advanced Search
Display Modes
Linear Mode Switch to Hybrid Mode Switch to Threaded Mode

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
bash script for pseudo-parallel usage of reconstructPar	kwardle	OpenFOAM Post-Processing	41	August 23, 2023 02:48
Job submit code in HPC for Workbench	beyonder1	CFX	0	January 6, 2016 11:18
Job submit code in HPC for Workbench	beyonder1	ANSYS	0	January 4, 2016 08:58
Slow Cases using HPC and PBS job scheduler	Alhasan	OpenFOAM Running, Solving & CFD	5	February 1, 2015 11:07
HPC manager job submission in powershell	ACmate	Siemens	0	January 30, 2015 04:11

June 5, 2017, 13:20		#2
evcelica Senior Member Erik Join Date: Feb 2011 Location: Earth (Land portion) Posts: 1,167 Rep Power: 23	You can set wall clock limit in CFX Pre. So the simulation will stop at say...71 hours. Can you then just restart the simulation using the previous results file as initial conditions? Therefore, just breaking it into two+ runs?