Intel® Math Kernel Library (Intel® MKL) Documentation

Intel® MKL Release Notes

Information about this release. Includes links to principal online information resources related to the release:

• Intel® MKL Main Product Page
• Intel® MKL 11.2 Release Notes (online version)
• Intel® MKL 11.2 Installation Guide
• Intel® MKL 11.2 System Requirements
• Intel® MKL 11.2 Getting Started
• Intel® MKL 11.2 Library Bug Fixes

Also includes the following information on:

• What's New in the Release
• Product Contents
• Obtaining Technical Support
• License Definitions

For details about the compiler and linking options, threading, and memory management with Intel® MKL, see Intel® MKL User's Guide.

Intel® MKL Reference Manual

Contains detailed descriptions of the functions and interfaces for all library domains:

• BLAS
• BLACS
• LAPACK
• ScaLAPACK
• PBLAS
• Sparse Solver
• Extended Eigensolver
• Vector Mathematical Functions (VML)
• Statistical Functions (VSL)
• Conventional DFTs and Cluster DFTs
• Partial Differential Equations support
• Non-Linear Optimization Problem Solvers
• Data Fitting Functions

For implementation-specific information about the library, see Intel® MKL Release Notes and Intel® MKL User's Guide.

Intel® MKL User's Guide

Provides Intel® MKL usage information in greater detail:

• getting started
• application compiling and linking depending on a particular platform and function domain
• building custom DLLs
• configuring the development environment
• coding mixed-language calls
• threading
• memory management
• improving performance from Intel® MKL functions

Intel® MKL Cookbook

Provides recipes for assembling Intel MKL routines for solving complex problems.

Tutorial: Using Intel® Math Kernel Library for Matrix Multiplication

• C Language
• Fortran Language

Demonstrates using Intel MKL to multiply matrices, measure the performance of matrix multiplication, and control threading.

Intel® Math Kernel Library Link Line Advisor tool (local version)

Returns the recommended line to use for linking and recommended compilation options given the Intel product version, operating system, processor architecture, and compiler, as well as other parameters.