ali kazempour wrote: > I didn't say that when I use 1 node each have 32 cores , > I don't have any problem with run of pw1.in and pw2.in with > nbnd=1024. But once I increase the number of proccessors to > 32 node each have 32 core this error happen.
...and what should one deduce from such obscure statements? The important pieces of information here are: - total number of MPI processes - number of processors in the "linear-algebra" group, as specified by the -ndiag or -northo keyword. This is a square integer, and should be set to whatever gives the fastest diagonalization time, not to the maximum number P. -- Paolo Giannozzi, Democritos and University of Udine, Italy
