[JLight]

Hi
I was wondering if anyone can suggest the compiler options to use with OpenFOAM-2.1.1 to squeeze out the maximum performance on a HPC. The HPC has Intel Xeon CPU E5-2670.

I initially compiled OF with:

1. Intel Compiler:
   

   icc --version
   icc (ICC) 12.1.3 20120212

   Options: -O3 -xAVX -no-prec-div -fp-model precise -fp-model source
2. GCC:
   

   gcc -v
   Using built-in specs.
   COLLECT_GCC=gcc
   COLLECT_LTO_WRAPPER=/usr/local/gcc/4.7.2/libexec/gcc/x86_64-unknown-linux-gnu/4.7.2/lto-wrapper
   Target: x86_64-unknown-linux-gnu
   Configured with: /usr/local/src/GCC/gcc-4.7.2/configure --prefix=/usr/local/gcc/4.7.2 --enable-shared --enable-threads=posix --enable-__cxa_atexit --enable-languages=c,c++,fortran --with-mpfr=/usr/local/mpfr/3.1.1/ --with-gmp=/usr/local/gmp/5.0.5/ --with-mpc=/usr/local/mpc/1.0/
   Thread model: posix
   gcc version 4.7.2 (GCC)

   Options: -march=corei7-avx -O3 -frounding-math -fsignaling-nans

My initial tests showed the one compiled with gcc was slightly faster when using the MotorBike Tute with simpleFoam.

Anyone have anything to share with regards to compiler performance with -O2/-O3 with gcc and/or intel or any other options (eg AVX)?

Cheers!

NJ

[wyldckat]

Hi NJ,

You can find some additional pointers on the comments and blog post here:
OpenFOAM with ICC - Intel's ICC is "allergic" to AMD CPUs

Best regards,
Bruno