Hi,<br>I have noted recently that I am able to obtain faster binaries of pw.x using the the OpenMP paralellism implemented in the Intel MKL libraries of version 10.xxx, than using MPICH, in the Intel cpus. Previously I had always gotten better performance using MPI. I would like to know of other experience on how to make the machines faster. Let me explain in more details.<br>
<br>Compiling using MPI means using mpif90 as linker and compiler, linking against mkl_ia32 or mkl_em64t, and using link flags -i-static -openmp. This is just the what appears in the make.sys after running configure in version 4cvs, <br>
<br>At runtime, I set<br>export OMP_NUM_THREADS=1<br>export MKL_NUM_THREADS=1<br>and run using <br>mpiexec -n $NCPUs pw.x <input >output<br>where NCPUs is the number of cores available in the system. <br><br>The second choice is <br>
./configure --disable-parallel<br><br>and at runtime <br>export OMP_NUM_THREADS=$NCPU<br>
export MKL_NUM_THREADS=$NCPU<br>
and run using <br>pw.x <input >output<br><br>I have tested it in Quadcores (NCPU=4) and with an old Dual Xeon B.C. (before cores) (NCPU=2).<br><br>Before April 2007, the first choice had always workes faster. After that, when I came to use the MKL 10.xxx, the second choice is working faster. I have found no significant difference between version 3.2.3 and 4cvs. <br>
<br>A special comment is for the FFT library. The MKL has a wrapper to the FFTW, that must be compiled after instalation (it is very easy). This creates additional libraries named like libfftw3xf_intel.a and libfftw2xf_intel.a<br>
This allows improves the performance in the second choice, specially with libfftw3xf_intel.a.<br><br>Using MPI, libfftw2xf_intel.a is as fast as using the FFTW source distributed with espresso, i.e., there is no gain in using libfftw2xf_intel.a. With libfftw3xf_intel.a and MPI, I have never been able to run pw.x succesfully, it just aborts. <br>
<br>I would like to hear of your experiences.<br> <br>Best regards<br>Eduardo Menendez<br>University of Chile<br><br>