Intel® Math Kernel Library Knowledge Base

Submit New Article

All Articles | Featured Articles | Compatibility | Download, Registration, Licensing | Errors and Diagnostics | Installation | Known Issues | Performance and Optimization | Sample Code | Software Defects | Technical Notes | Tips and Techniques


Title Modified Date
Performance slow down when dynamically linking with Intel MKL
Performance slows down with dynamic linkage
Author: Chao Y (Intel)
Type: Performance and Optimization
 11/24/2009
Intel MKL Threaded Functions
threaded function included in Intel MKL
Author: Chao Y (Intel)
Type: Performance and Optimization
 11/24/2009
Different parallelization techniques and Intel® MKL FFT
This article is to describe various techniques for threading which can be used by the user and how to set the right values for environment variables or DFTI parameters.
Author: Vipin Kumar E K (Intel)
Type: Performance and Optimization
 11/22/2009
Build CP2K using Intel Fortran compiler Professional Edition
This article show how to build CP2K for Intel64 platform, using Intel Fortran compiler Professional Edition version 11.0.083 / 11.1.059, and Intel Math Kernel Library (MKL) with FFTW 3.x support
Author: Hao Jiang (Intel)
Type: Performance and Optimization
 11/03/2009
Running The HPL Benchmark Over Intel MPI
This is a step by step procedure on how to run the High Performance Linpack (HPL) benchmark on a Linux cluster using Intel-MPI. This was done on a Linux cluster of 128 nodes running Intel’s Nehalem processor 2.93 MHz with 12GB of RAM on each node.
Author: Mohamad Sindi
Type: Performance and Optimization
 09/28/2009
Enabling Intel® Advanced Vector Extensions (Intel® AVX) optimizations in Intel® MKL
Intel® MKL 10.2 has early support of Intel® AVX and includes special optimizations for several BLAS, DFT and VML functions.
Author: Vipin Kumar E K (Intel)
Type: Performance and Optimization
 08/10/2009
Intel® MKL Threaded 1D FFTs
This document describes the cases for which the Intel MKL 10.2 and later 1D complex-to-complex FFTs are threaded.
Author: Dmitry Baksheev (Intel),AmandaS (Intel)
Type: Performance and Optimization
 07/17/2009
New in Intel® MKL 10.2
A list of performance improvements, new features, interface improvements in Intel® MKL 10.2 and its updates
Author: Todd Rosenquist (Intel)
Type: Performance and Optimization
 06/23/2009
Setting thread affinity on SMT or HT enabled systems for better performance
For SMT (Simultaneou Mutli Threading) or HT (Hyper Threading) enabled systems, for. e.g: "Nehalem" based machines to get best MKL performance we recommend you to set thread affinity.
Author: Vipin Kumar E K (Intel)
Type: Performance and Optimization
 06/22/2009
Intel® MKL LAPACK 3.2
Intel® MKL 10.2 introduces compatibility with LAPACK 3.2
Author: Todd Rosenquist (Intel)
Type: Performance and Optimization
 06/22/2009
Enabling parallelization in reordering and symbolic factorization steps in PARDISO and DSS
Reordering and symbolic factorizations steps in PARDISO and DSS are threaded Intel MKL 10.2 version onwards. This articles explains how to enable the parallel version of the above steps.
Author: Vipin Kumar E K (Intel)
Type: Performance and Optimization
 05/05/2009
Threaded BLAS Level 1 and Level 2 functions are available in Intel MKL 10.2 on Intel Xeon 5500 series (Nehalem) processors
Threaded version BLAS level 1 and level 2 functions available in Intel MKL 10.2 for Intel Xeon 5500 (Nehalem) processors.
Author: Vipin Kumar E K (Intel)
Type: Performance and Optimization
 05/03/2009
Intel® Xeon 5500 (Nehalem) processor optimization in Intel® MKL
Intel® MKL 10.1 supports Intel® Xeon 5500 processors (codenamed "Nehalem").
Author: Vipin Kumar E K (Intel)
Type: Performance and Optimization
 04/02/2009