On Feb 5, 2013, at 22:12 , pascal boulet wrote: > I guess the best thing to do is to parallelize the calculation on > the bands
"band" parallelization is currently implemented only for some specific cases. Same for "image" parallelization. There is no quick and easy answer to your question. Your calculation is relatively small, so you should likely exploit k-point (-npool 4) and FFT parallelization, using a few processors. How many it depends upon the size of your FFT grids. Here some explanations: http://www.fisica.uniud.it/%7Egiannozz/QE-Tutorial/tutorial_para.pdf See also the general user guide, chapter 3: http://www.quantum-espresso.org/wp-content/uploads/Doc/user_guide/ and the user guide for PWscf, sec. 4.4 and 4.5: http://www.quantum-espresso.org/wp-content/uploads/Doc/pw_user_guide/ P. --- Paolo Giannozzi, Dept of Chemistry&Physics&Environment, Univ. Udine, via delle Scienze 208, 33100 Udine, Italy Phone +39-0432-558216, fax +39-0432-558222
