[Wien] parallel calc. without NFS

[Torsten Andersen]

Steven Homolya wrote:

>On Wed, 9 Jul 2003, Jeff Spirko wrote:
>
>  
>
>>You might try smoothing out some of that I/O.  Using GNU versions of
>>sed and awk may help. Or, piping the input our output through cat or
>>dd (with some blocksize ~10k) may force that I/O to be buffered.
>>
>>    
>>
>
>Thanks. I'll try that (though I'm told blocksize needs to be >= 1M).
>
>Do any of the fortran sources read/write large files small segments at a
>time?
>
>  
>
Yes - lapw1, lapwso, lapw2, and several of the analysis programs reads 
or writes data for one k-point at a time to/from the vector files (they 
are the large files present), but it is always serial (no random access 
is needed) so that should not in itself cause any problems. Your 
operating system should do the buffering automatically.

Best regards,
Torsten Andersen.


-- 
Dr. Torsten Andersen
Department of Physics, Condensed Matter Theory Group, Uppsala University
UU-WWW: http://www.fysik4.fysik.uu.se/    TA-WWW: http://deep.at/myspace