[Pw_forum] Re: about output wavefunction, charge density and potential

stewart at cnf.cornell.edu stewart at cnf.cornell.edu
Wed Apr 5 15:40:47 CEST 2006


Hi J G, 

  Try setting the option wf_collect = .true. in the &control section.  This 
will collect the wave functions from all the processors and store everything 
in a save file for restarting (see INPUT_PW for details).  This will 
increase traffic on the network some but I believe the collection is only 
done at the end of the calculation. 

Best regards, 

Derek 

 

J G Che writes: 

> If so, are there any options, with which all files in scratch will be saved on user's disk once before exiting calculation? Otherwise, it cannot be re-started from scratch, since PBS distributes a job into avaiable nodes, not into the defined nodes as same as the last calculation. Or it can only be run in a small cluster which is only for one person.  
> 
> JG
> ----- Original Message ----- 
> From: "Paolo Giannozzi" <giannozz at nest.sns.it>
> To: <pw_forum at pwscf.org>
> Sent: Tuesday, April 04, 2006 5:00 PM
> Subject: Re: [Pw_forum] about output wavefunction, charge density and potential 
> 
> 
>> On Tuesday 04 April 2006 09:56, J G Che wrote: 
>> 
>> > However, now there are many abinit packages based plane-waves
>> > which don't use disk but RAM to store wavefunctions, meaning it
>> > could be at least as a option. Otherwise, such an often writing on 
>> > disk not only wastes too much time, but also leads to NFS crashing 
>> 
>> one should never use NFS-mounted filesystem for I/O. Use locally
>> mounted scratch disks, or a parallel file system (GPFS) if you have one. 
>> 
>> Also note that modern operating system tend to keep data in memory
>> as long as possible. Data to disk is actually written when no more RAM
>> for caching is available. 
>> 
>> Anyway: I agree that it is useful in many cases to keep everything 
>> in memory, but this requires some work 
>> 
>> Paolo
>> -- 
>> Paolo Giannozzi             e-mail:  giannozz at nest.sns.it
>> Scuola Normale Superiore    Phone:   +39/050-509876, Fax:-563513 
>> Piazza dei Cavalieri 7      I-56126 Pisa, Italy
>> _______________________________________________
>> Pw_forum mailing list
>> Pw_forum at pwscf.org
>> http://www.democritos.it/mailman/listinfo/pw_forum
> _______________________________________________
> Pw_forum mailing list
> Pw_forum at pwscf.org
> http://www.democritos.it/mailman/listinfo/pw_forum
 


################################
Derek Stewart, Ph. D.
Scientific Computation Associate
250 Duffield Hall
Cornell Nanoscale Facility (CNF)
Ithaca, NY 14853
stewart (at) cnf.cornell.edu
(607) 255-2856 




More information about the Pw_forum mailing list