I do not know what happened with my reply, but here it is again; Yes, you are right; if after symmetry you have N k-points, lets say for example 53 k-points, and you use 5 cores, then the machine will run 10 k-points in each core and after this it will 1 k-point in 3 cores. But as Laurence Marks says; you can edit the .machine* files and add the extra k-points in 3 of the .machine* files, and delete the extra .machine* files... (I have not tried this last possibility)
________________________________________ De: wien-boun...@zeus.theochem.tuwien.ac.at <wien-boun...@zeus.theochem.tuwien.ac.at> en nombre de Ридный Ярослав Максимович <yaros...@physics.susu.ac.ru> Enviado: martes, 7 de abril de 2015 06:01 a. m. Para: wien@zeus.theochem.tuwien.ac.at Asunto: [Wien] Parallel computing Hi, I have a question. I do my calculations on supercomputer. It is located on one node 12 cores (processors). If the number of k-points higher than the number of processors, as happens parallelization The first stage of parallelization goes on to-point to different processors. The second stage goes parallelization rate of 1 k-points on multiple processors. Do I understand correctly or not. _______________________________________________ Wien mailing list Wien@zeus.theochem.tuwien.ac.at http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien SEARCH the MAILING-LIST at: http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/index.html _______________________________________________ Wien mailing list Wien@zeus.theochem.tuwien.ac.at http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien SEARCH the MAILING-LIST at: http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/index.html
