Optimize Intel AVX with Math Kernel Library Share your comment!

The Intel® Math Kernel Library (Intel® MKL) provides broad functionality for scientific and engineering use. These include:

  • Linear Algebra – Basic Linear Algebra Subprograms (BLAS), LAPACK, ScaLAPACK, Sparse BLAS, Iterative Sparse Solvers, Preconditioners, Direct Sparse Solvers (PARDISO)
  • FFTs – Both sequential and cluster FFTs
  • Statistics – Vector Statics Library (VSL) and random number generators
  • Vector Math – Vector Math Library (VML)
  • PDEs – Poisson, Helmholtz solvers, trigonometric transforms
  • Optimization – Trust Region Solvers

 This technical paper provides a brief outline of Intel MKL, then discusses generic performance-related enhancements with regard to DGEMM and new Intel® Advanced Vector Extensions (Intel® AVX) instructions. 

Read the full paper here.

Posted on January 7, 2013 by Gregory Henry, Principal Engineer, Intel®