Intel® Xeon Phi optimizations in Intel MKL

The following components of Intel® MKL 11.0.1 and higher are tuned for the Intel® Xeon Phi Architecture:

  • Several BLAS (level 1, 2, and 3)
  • Sparse BLAS
  • LAPACK routines.
  • Vector Math Library (VML)
  • All the Vector Statistical Library (VSL) routines including random number generators (RNG).
  • Fast Fourier transforms. 
    1. Please refer the FFT tuning article for Intel Xeon Phi for more details.

Here are some of the Automatic Offload enabled functions list

Many of the other routines which uses ?gemm also get performance benefits due to ?gemm optimizations.

Please refer other articles related to Intel MKL on Intel Xeon Phi at Intel® Math Kernel Library on the Intel® Xeon Phi™ Coprocessor

For more complete information about compiler optimizations, see our Optimization Notice.