[gmx-developers] CVS implementation of REMD

[Anton Feenstra]

Ken Rotondi wrote:

> On May 25, 2005, at 11:19 AM, Anton Feenstra wrote:
> 
>> Ken Rotondi wrote:
>>
>>> Then I ran mdrun:
>>> mdrun -v -multi -np 2 -replex 20 -s topol.tpr
>>
>>
>> Perhaps -np 2 and -replex 20 'bite' each other. From what I gather 
>> from the other mails, 'mdrun -v -multi -replex 2' could work.
>>
>> -- 
>> Groetjes,
>>
>> Anton
> 
> 
> This is rather confusing. I thought the -np flag was for the number of 
> processes running (i.e. the number of replicas in a REMD run) while the 
> -replex flag defines the number of integration steps that pass between 
> attempts at exchange.

The way I understand it, '-np' defines the number of processors to use 
for one single simulation. In your case, you are actually performing 
multiple simulations at the same time. Have you tried mdrun without the 
'-np 2'?
It may be confusing, but if it works you can get on with your work ;-)

-- 
Groetjes,

Anton

* NOTE: New Phone & Fax numbers (below) *

  _____________ _______________________________________________________
|             |                                                       |
|  _   _  ___,| K. Anton Feenstra                                     |
| / \ / \'| | | Dept. of Pharmacochem. - Vrije Universiteit Amsterdam |
|(   |   )| | | De Boelelaan 1083 - 1081 HV Amsterdam - Netherlands   |
| \_/ \_/ | | | Tel: +31 20 59 87608 - Fax: +31 20 59 87610           |
|             | Feenstra at few.vu.nl - www.few.vu.nl/~feenstra/         |
|             | "If You See Me Getting High, Knock Me Down" (RHCP)    |
|_____________|_______________________________________________________|