Intel® MKL 11.3 Beta (released in April 2015) contains significant performance and scalability improvements for the direct sparse solver (a.k.a. Intel MKL PARDISO), on SMP systems. These improvements particularly benefit the Intel Xeon Phi coprocessors and Intel Xeon processors with large core counts. As an example, the chart below shows a 1.7x to 2.5x speedup of Intel MKL 11.3 Beta over Intel MKL 11.2, when using the PARDISO to solve various sparse matrices on an Intel Xeon Phi coprocessor with 61 cores.
Dynamic Program Slicing is a dynamic program analysis technique that given a slicing criterion (line number, variable,..) finds all statements in the program affecting (if backward sliced) or affected by (if forward sliced) the slicing criterion for a specific execution.
The code samples for the webinar "Further Vectorization Features of the Intel Compiler" given on 4/7/2015 are attached below.
Here are some examples of command lines that may be used to build them. This list is not intended to be complete or to be a tutorial as such, just a guide to things to try. It uses Linux* switch syntax; Windows* equivalents are closely similar.
See the presentation for more detail; slides and video will be posted separately and later.
icpc -c -qopt-report-phase=vec -qopt-report=3 no_stl.cpp