It is, but limiting the RAM usage and using only one core so that one can run jobs in parallel will likely be much slower than allowing it to use all cores with as much RAM as possible.

Peace,

Matt.

From: <House>, Matthew <[email protected]>
Date: Tuesday, November 11, 2014 at 12:27 PM
To: Oscar Miranda Dominguez <[email protected]>, Matt Glasser <[email protected]>
Cc: "[email protected]" <[email protected]>
Subject: RE: [HCP-Users] How to limit the number of cores when wb_command is running?

Hi, Oscar.  Speaking briefly with our local High Performance Computing admin, he also suggesting using the OMP_NUM_THREADS environmental variable (assuming wb_command is using OpenMP, which I’m not 100% certain of).

 

~Matt

 

From: [email protected] [mailto:[email protected]] On Behalf Of Oscar Miranda Dominguez
Sent: Tuesday, November 11, 2014 12:25 PM
To: Glasser, Matthew
Cc: [email protected]
Subject: Re: [HCP-Users] How to limit the number of cores when wb_command is running?

 

Ok.

I’ll try that.

 

I’ll also give another try to the // submission setting this variable before running the jobs in //:

export OMP_NUM_THREADS=1

 

Thanks and best regards

OMD

 

From: Glasser, Matthew [mailto:[email protected]]
Sent: Tuesday, November 11, 2014 10:19 AM
To: Oscar Miranda Dominguez
Cc:
[email protected]
Subject: Re: [HCP-Users] How to limit the number of cores when wb_command is running?

 

Perhaps you should have one job at a time requesting all cores (and allow it to use more RAM)?  That will likely be more efficient than running multiple jobs in parallel.

 

Peace,

 

Matt.

 

From: Oscar Miranda Dominguez <[email protected]>
Date: Tuesday, November 11, 2014 at 12:15 PM
To: "
[email protected]" <[email protected]>
Subject: [HCP-Users] How to limit the number of cores when wb_command is running?

 

Hi

 

Is it possible to control the number of cores when using wb_command?

I was using wb_command -cifti-correlation to calculate 11 connectivity matrices using a grid engine. However, all the jobs were killed since each was making use of all the available cores.

The documentation specify how to control the memory but I could not find how to specify the number of cores

 

Thanks!

 

Oscar Miranda-Dominguez

Postdoctoral Researcher

Fair Neuroimaging Lab

Phone: 503.418.1894  Fax: 503.494.9988

 

Oregon Health and Science University

Department of Behavioral Neuroscience

3181 SW Sam Jackson Park Road, L470

Portland, OR  97239

http://www.ohsu.edu/fair-lab

 

_______________________________________________
HCP-Users mailing list
[email protected]
http://lists.humanconnectome.org/mailman/listinfo/hcp-users

 


The materials in this message are private and may contain Protected Healthcare Information or other information of a sensitive nature. If you are not the intended recipient, be advised that any unauthorized use, disclosure, copying or the taking of any action in reliance on the contents of this information is strictly prohibited. If you have received this email in error, please immediately notify the sender via telephone or return mail.

_______________________________________________
HCP-Users mailing list
[email protected]
http://lists.humanconnectome.org/mailman/listinfo/hcp-users




The material in this message is private and may contain Protected Healthcare Information (PHI). If you are not the intended recipient, be advised that any unauthorized use, disclosure, copying or the taking of any action in reliance on the contents of this information is strictly prohibited. If you have received this email in error, please immediately notify the sender via telephone or return mail.

 


The materials in this message are private and may contain Protected Healthcare Information or other information of a sensitive nature. If you are not the intended recipient, be advised that any unauthorized use, disclosure, copying or the taking of any action in reliance on the contents of this information is strictly prohibited. If you have received this email in error, please immediately notify the sender via telephone or return mail.

_______________________________________________
HCP-Users mailing list
[email protected]
http://lists.humanconnectome.org/mailman/listinfo/hcp-users

Reply via email to