Thank you for your response, Stefano. I appreciate it. o.
On Mon, Apr 11, 2011 at 10:17 AM, Stefano de Gironcoli <degironc at sissa.it> wrote: > because the code estimate that the communication overhead would exceed > the gain due to the distribution of the operation across the processors. > This apply to the subspace diagonalization, the rest of the run is parallel > > stefano > > On 04/11/2011 08:52 AM, Omololu Akin-Ojo wrote: >> Hi, >> >> I was aiming to run QE in parallel but I see the following message in >> my output file; what am i doing wrong? >> >> ? ? ? Subspace diagonalization in iterative solution of the eigenvalue >> problem: >> ? ? ? a serial algorithm will be used >> >> Why is QE using a serial algorithm? I am using openmpi and QE >> recognizes the number of processors correctly: >> ? ? ? Parallel version (MPI), running on ? ? 8 processors >> ? ? ? R& ?G space division: ?proc/pool = ? ?8 >> >> I could give more info if needed. >> >> Thanks in advance for any help. >> >> o. >> >> Postdoctoral fellow, >> CMSP section, >> ICTP, Italy. >> >> >> > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > -- ***************** Seek GOD! *************
