[phenixbb] meaning of nproc in mr_rosetta

Thomas C. Terwilliger terwilliger at lanl.gov
Tue Jun 7 05:42:46 PDT 2011


Hi Miguel,

Yes, in your case it should run the first 5 models of each hhr file...a
total of 10 search models.  Let me know if a more recent phenix fixes the
nproc problem (in your case if you divide nproc by 10 it should do what
you want...).

Let me know of other problems!

All the best,
Tom T

>> Le 07/06/11 14:09, Thomas C. Terwilliger a écrit :
>>> Hi Miguel,
>>>
>>> Normally nproc is how many processors are to be used at any one time by
>>> phenix.mr_rosetta.
>>>
>>> There was a problem in versions up to dev-767 where nproc was not
>>> divided
>>> by the number of search models you supply, so each search model got
>>> nproc
>>> processors devoted to it... That should be fixed, so if you are using a
>>> more recent version than that and you get more processors in use than
>>> requested, please let me know.
>>>
>>> All the best,
>>> Tom T
>>>
>>
>> Hi Tom,
>>
>> Thank you for your quick answer. We are using dev-764, you go too fast!
>> :-)
>>
>> I didn't supply any models directly, just two hhr files, so I guess the
>> first five models from each hhr file are used? Add to this that the
>> space group has an enantiomorph (and it seems to me that both are tested
>> even if use_all_plausible_sg is set to False) and that I requested
>> prerefining. For the time being is not sending more than ~ 170 jobs at a
>> time, which should be fine.
>>
>> I will let you know if newer versions behave differently.
>>
>> Best regards,
>>
>>
>> --
>> Miguel
>>
>> Architecture et Fonction des Macromolécules Biologiques (UMR6098)
>> CNRS, Universités d'Aix-Marseille I & II
>> Case 932, 163 Avenue de Luminy, 13288 Marseille cedex 9, France
>> Tel: +33(0) 491 82 55 93
>> Fax: +33(0) 491 26 67 20
>> mailto:miguel.ortiz-lombardia at afmb.univ-mrs.fr
>> http://www.afmb.univ-mrs.fr/Miguel-Ortiz-Lombardia
>>
>> --
>> This message has been scanned for viruses and
>> dangerous content by MailScanner, and is
>> believed to be clean.
>>
>> _______________________________________________
>> phenixbb mailing list
>> phenixbb at phenix-online.org
>> http://phenix-online.org/mailman/listinfo/phenixbb
>>



More information about the phenixbb mailing list