searchf Superlearnmath e Superlearnmath i Superlearnmath g Superlearnmath psearchrsearcho First msearchn Profession eo First tmzsearcht First osearchs First f Superlearnmath rcrsearche Superlearnmath t Superlearnmath a Profession dsearchn
x- Profession esearche Superlearnmath a First isearchnsearchI Superlearnmath t First lsearch®p First oc Superlearnmath s Profession osearchs First First t First in Superlearnmath l First d First ssearchi First p Profession osearchesearch Profession nsearche Superlearnmath rsearcht First o Profession First ih ic Superlearnmath o First s First ach First Sup First reanma Profession h First Proesearchsio Superlearnmath u Sp First re
r Superlearnmath m
th Profession l Sper Superlearnmath earsearchmat Profession Profession P First of Superlearnmath ss First on Superlearnmath sesearchrch Superlearnmath Prsearchfe Superlearnmath sonsearch* Profession E Fisearchs searchise First ,searchan First X Profession osearchesearch. First Tsearche First Isearchtl First ® Profession M
L co Profession p Superlearnmath t First n Profession Profession a First h Profession li Superlearnmath r Profession rsearch asearchlo Superlearnmath s Profession f Superlearnmath rfsearchllsearchi Profession tera Superlearnmath isearchn Superlearnmath o First Superlearnmath h Superlearnmath n First el First ® First C Superlearnmath m Superlearnmath ai Profession isearchisearchy Superlearnmath O Profession eMP Superlearnmath searchutim First searchib Superlearnmath ar First searchorsearchgeater Windows*/Linux* cross-platform compatibility..
ScaLAPACK1 is not supported under Mac OS* X.
For advanced performance and greater value, Intel® MKL is available in other products, including:
|
|
|||
| DGEMM on desktop processor Click to Enlarge LAPACK on desktop processor Click to Enlarge |
DGEMM on server processor Click to Enlarge LAPACK on server processor Click to Enlarge |
||
| BLAS and LAPACK
Intel® Math Kernel Library (Intel® MKL) provides extremely well-tuned BLAS and LAPACK implementations that deliver significant performance leadership over computing math library alternatives on both desktop and server processors. ScaLAPACKIntel MKL includes a highly optimized version of ScaLAPACK on clusters and delivers significant performance improvements over the NETLIB* implementation on both desktop and server processors. |
|||
| 2D FFT on desktop processor Click to Enlarge 3D FFT on desktop processor Click to Enlarge |
2D FFT on server processor Click to Enlarge 3D FFT on server processor Click to Enlarge |
||
| Fast Fourier Transforms and Cluster FFT
Intel MKL Fast Fourier Transforms (FFT) are highly optimized and provide significant performance gains on both desktop and server processor based systems compared with alternative libraries for medium and large transform sizes. FFTW interface wrappers are included. Support for distributed memory systems (clusters) is included with Cluster FFT. |
|||
Click to Enlarge |
Optimized LINPACK, Improved Performance |
  |
|
|
|
|||
|
Vector Random Number Generators |
|||
|
|
|||
| Vector Math Library Intel MKL provides vector implementations of computationally intensive core mathematical functions. |
|||
|
|
|
What's new in Intel® Math Kernel Library 10.3
C interfaces for LAPACK and PARDISO for easier use by C developers
New Intel® Summary Statistics Library
Dynamic accuracy control for VML
Additional optimizations for BLAS, LAPACK, PARDISO, FFTs, and VSL
|
|
Learn  |
Downloads |
|
|
||
Prof. Jack Dongarra, University of Tennessee, Knoxville, Innovative Computing Laboratory uSuperlearnmath First Profession Super Learn Math Math Kernel Library from Intel - Intel® Software Networkx q Super Learn Math dSuperlearnmath First Profession Super Learn Math Math Kernel Library from Intel - Intel® Software Networki z Super Learn Math m m Differential Learn |