Documentation on the parallelization of BSE ?

Deals with issues related to computation of optical spectra in reciprocal space: RPA, TDDFT, local field effects.

Moderators: Davide Sangalli, andrea.ferretti, myrta gruning, andrea marini, Daniele Varsano, Conor Hogan

Post Reply
Flex
Posts: 37
Joined: Fri Mar 25, 2016 4:21 pm

Documentation on the parallelization of BSE ?

Post by Flex » Tue Jun 20, 2017 1:54 pm

Hello,

This may be a dumb question, but I don't seem to find any documentation about the parallel variables for BSE. Moreover, when I type something like
¨yambo -o b -k sex -y d -V par¨, I don't get any parallel variables in the generated input file.
I know they exist from this external page : http://www.attaccalite.com/yambo-input- ... explained/
But this may be outdated, or not supported, so I'm not sure if I should try.

My calculations usually need a lot of optimisation. These variables, if they still work properly, and some reference on how to use them, would be highly useful for me.

Thanks in advance
Thierry Clette
Student at Université Libre de Bruxelles, Belgium

User avatar
Daniele Varsano
Posts: 4231
Joined: Tue Mar 17, 2009 2:23 pm
Contact:

Re: Documentation on the parallelization of BSE ?

Post by Daniele Varsano » Tue Jun 20, 2017 3:46 pm

Dear Thierry,
The name of the variable is the same you have found in the page of Claudio Attaccalite.
I do not know why the -V par is not working, we will check it.
If you have a k point sampling the suggestion is to parallelize over "k" as it is the most effective, then "t' as "eh" it is not efficient.
If you have an isolated system, gamma only, then "eh" parallelism is suggested then "t".
Best,
Daniele
Dr. Daniele Varsano
S3-CNR Institute of Nanoscience and MaX Center, Italy
MaX - Materials design at the Exascale
http://www.nano.cnr.it
http://www.max-centre.eu/

Post Reply