Intel mkl basics (1) Understanding MKL and MKL Resources

Intel MKL: Intel math kernel library, Intel Core mathematical function library.

The following is a product introduction page for MKL on the intel website, which can be used for preliminary understanding of MKL:

Intel MKL product introduction:

Http://software.intel.com/zh-cn/articles/intel-mkl/ (http://software.intel.com/en-us/articles/intel-mkl)

Links to Intel MKL-related documents (provide MKL-related documents ):

Http://software.intel.com/en-us/articles/intel-math-kernel-library-documentation/

Intel MKL Reference Manual (reference manual) (can be used to query descriptions and usage of all MKL functions ):

Http://software.intel.com/sites/products/documentation/hpc/mkl/mklman/index.htm

The following is an overview of mklon http://software.intel.com/zh-cn/articles/intel-mkl:

Intel's core mathematical function library is a set of highly optimized and widely-threaded mathematical routines designed for scientific, engineering, financial, and other applications that require extreme performance. Core mathematical functions include Blas, LAPACK, scalapack1, sparse matrix calculator, Fast Fourier transformation, vector mathematics, and other functions.
It provides performance optimization for current and next-generation Intel processors, including better integration with Microsoft Visual Studio, eclipse, and xcode. Intel MKL supports full integration with Intel-compatible OpenMP Runtime Library for better cross-platform compatibility between Windows and Linux.
(The Mac OS X operating system does not support scalapack .)

To sum up, we can get the following characteristics of MKL (http://www3.intel.com/cd/software/products/apac/zho/329191.htm ):

(1) Interface Support: Intel MKL is a highly optimized and threaded function library that provides C and FORTRAN interfaces.

(2) processor support: it can provide performance optimization for current and next-generation processors. It supports all processors compatible with Intel processors. Note: MKL runs a more running processor environment and automatically checks the processor during running to run different optimized versions of programs for different processors, so as to ensure that it can obtain better performance on the running processor, it is possible that the same program using MKL may have different performance on different processors, because MKL will detect different processors and optimize them as much as possible.

(3) platform and tool support: supports mainstream operating systems (such as Windows, Linux, and Mac OS) and integrates with mainstream development tools (such as Vs, Eclipse, xcode, and GCC.

(4) multi-core multi-thread expansion performance and thread security:
The built-in Parallel Processing Mechanism automatically delivers excellent scalability on multi-core and multi-processor. All MKL functions are thread-safe. It also provides a non-threaded sequential MKL.

(5) functions:

Linear Algebra (BLAs and LAPACK)
Linear Algebra-scalapack
Linear Algebra-sparse matrix Calculator
Fast Fourier Transform (FFT)
Vector Math Library
Vector random number generator
Linpack Performance Index Evaluation

.

For more information, see related documents.