Intel® MKL


Intel® Math Kernel Library
Benefit from performance optimizations for current and future Intel® processors with math routines for science, engineering, and financial applications that require maximum performance.


Power science, engineering and financial applications with this highly optimized library


Intel® Math Kernel Library (Intel® MKL) is a library of highly optimized, extensively threaded math routines for science, engineering, and financial applications that require maximum performance. Core math functions include BLAS, LAPACK, ScaLAPACK, Sparse Solvers, Fast Fourier Transforms, Vector Math, and more. Offering performance optimizations for current and next-generation Intel® processors, it includes improved integration with Microsoft Visual Studio*, Eclipse*, and XCode*. Intel MKL allows for full integration of the Intel Compatibility OpenMP* run-time library for greater Windows*/Linux* cross-platform compatibility.




 Benefits:
  • Outstanding performance - multicore and multiprocessor ready
  • Automatic parallelization
  • Standard APIs in C and Fortran
  • Royalty free redistribution
  • World-class technical support, knowledge base, and active Intel MKL forum

BLAS and LAPACK
Intel MKL provides extremely well-tuned BLAS and LAPACK implementations that deliver significant performance leadership over alternative math libraries.

ScaLAPACK
Intel MKL includes a highly optimized version of ScaLAPACK on clusters and delivers significant performance improvements over the NETLIB* implementation.

Fast Fourier Transforms and Cluster FFT
Intel MKL Fast Fourier Transforms are highly optimized and provide significant performance gains over alternative libraries for medium and large transform sizes. FFTW interface wrappers are included. Support for distributed memory systems (clusters) is included with Cluster FFT.

Vector Random Number Generators
Intel MKL Vector Statistical Library (VSL) is a collection of 9 random number generators and 22 probability distributions that deliver significant performance improvements in physics, chemistry, and financial analysis.

Sparse Solvers
The library includes both direct and iterative sparse solvers:

    Direct solvers - PARDISO: A threaded, high-performance, memory efficient solver for large sparse linear systems of equations. Includes support for out-of-core memory.

    Iterative solvers - FGMRES* and Conjugate Gradient Solvers: FMGRES adds the capability to solve general sparse systems of linear equations while the Conjugate Gradient solver solves symmetric positive-definite systems

Vector Math Library
Intel MKL provides vector implementations of computationally intensive core mathematical functions.

Optimized LINPACK benchmark
The Intel MKL package includes an optimized implementation of the LINPACK benchmark which is easy to run on any Intel architecture platform and provides the best performance on the latest Intel processors, getting close to the maximum Gflops supported by the underlying platform.


To learn more about Intel Math Kernel Library, download the product brief › or product in-depth ›
 
Prof. Jack Dongarra, University of Tennessee, Knoxville, Innovative Computing Laboratory
"University of Tennessee, Knoxville: “The Intel® Math Kernel Library is indispensable for any high performance computer user on x86 platforms. It provides a rich, highly optimized collection of math routines, including our own BLAS, LAPACK, and ScaLAPACK in addition to other basic functions such as sparse matrix operations and FFTs. Outstanding performance is achieved on both multicore and multiprocessor systems.”."
 
Dr Antoine Petitet, Ph.D, HPC Lead , ESI Computational Structural Mechanics Group
"ESI Computational Structural Mechanics Group: “PAM-CRASH and PAM-STAMP rely on the performance of the Intel® Math Kernel Library 10.0. We are delighted with the results in terms of both memory use and SMP performance."
 
Gilles Artaud, Head of Quantitative Research, Risk and Permanent Control, CALYON
"CALYON counter party risk is computed using complex numerical algorithms, distributed on a cluster of hundreds of Intel processors. We are constantly facing performance issues. The Intel® MKL includes vectors and matrixes calculation, FFT, random generators, and many other algorithms which allowed substantial performance improvements, sometimes up to a hundred times, and reduced development costs."
 
Chris Reid, Vice President of Marketing, ANSYS, Inc.
"Intel MKL helps ANSYS achieve excellent performance on Intel processors and has powered our engineering simulation software for more than 10 years. Intel multi-core processors coupled with the Intel MKL library help us deliver high performance, with engineered scalability from workstations to server systems. The continued optimization of Intel MKL ensures the best performance for ANSYS software users on the latest generation Intel processors."
 
Alistair Downie, Paradigm Geophysical
"Paradigm use Intel® MKL in our technical solutions for our oil and gas customers because it provides outstanding performance. Our HPC algorithms rely on Intel MKL being tuned, thread safe and multicore aware across Intel’s processor families. Using Intel MKL enables us to focus our time and effort on our solutions, yet still deliver optimal performance across a range of systems, from laptops and workstations to HPC clusters."
Intel® Math Kernel Library Support

Browse the Intel® Math Kernel Library Knowledge Base



Ask Experts Online


Search and post a new question to the Intel® Software Network Forum for Intel®Math Kernel Library. The forums allow you to search a growing archive of technical questions and answers from both Intel experts and our developer community. If an answer can't be found in Search, you can ask a new question. An Intel® Software Development Products Registration Center login will allow you to participate in the Intel® Software Network User Forums without any additional registration.



Get Help from Intel


If you do not find what you need above:
Submit Software Tools bugs through Intel® Premier Support. Registration is required. Click here for more information.