Numerical Computation on Intel® Xeon Phi™ Coprocessors Using the Intel® Compilers and Math Libraries
Intervenant(s) : Marius Cornea (Intel)
Intel® Xeon Phi™ coprocessors (the first implementation of the Intel® MIC Architecture) are described briefly, with emphasis on the support they offer for floating-point computations.
Usage of Intel® Compilers for numerical computation on Intel® Xeon® processors and Intel® Xeon Phi™ coprocessors, along with programming models for Intel® Xeon Phi™ are presented.
Mathematical libraries (in particular Intel® MKL) and their usage on Intel® Xeon Phi™ are overviewed, followed by a selection of performance numbers.
Information about the next generation of Intel® Xeon Phi™ processors and coprocessors is provided in the last part of the presentation.