Tips and techniques on using the Intel® Compilers to maximize your application performance.
The purpose of this document is to help developers determine which FFT, Intel® MKL or Intel® IPP is best suited for their application.
Vectorization is one of many optimizations that are enabled by default in the latest Intel compilers. In order to be vectorized, loops must obey certain conditions, listed below. Some additional ways to help the compiler to vectorize loops are described.
A toolkit that gives 6 Steps to Increase Performance Through Vectorization in Your Application
Intel Vectorization Toolkit: 3. Vectorization report for loop candidates
Which applications are most likely to benefit from recompilation for Intel® Advanced Vector Extensions (Intel® AVX)?Applications containing vectorizable, floating-point loops or calls to performance libraries are the most likely to see significant performance gains from rebuilding for the Intel® Advanced Vector Extensions (Intel® AVX)
Compiler Methodology for Intel® MIC Architecture
Examples of vectorizing Fortran applications
Vectorization Essentials: Vectorizing the outer loop can be profitable
Vectorization Essentials: Efficient vectorization involves making full use of the vector-hardware in the kernel-vector loop.