Developer Reference


Notes for Intel® Math Kernel Library Vector Statistics

Vector Statistics (VS) is a component of Intel® Math Kernel Library (Intel® MKL). VS performs a range of statistical computations, including random number generation, summary statistics and convolution calculations.
This document includes an overview, a usage model and testing results of random number generators included in VS.
BRNG Tests Description
BRNG Tests Description contains detailed descriptions and passing criteria of the empirical tests implemented in Vector Statistics.
Basic Random Number Generator Properties and Testing Results
Basic Random Number Generator Properties and Testing Results provides a brief description and properties of each Basic Random Number Generator (BRNG) included in Vector Statistics, BRNG reference information, and BRNG empirical testing results.
Distribution Random Number Generators
Continuous Distribution Random Number Generators and Discrete Distribution Random Number Generators briefly describe the methods used to generate random number sequences from given statistical distributions.
Random Streams and RNGs in Parallel Computation
Random Streams and RNGs in Parallel Computation provides insight into the different aspects of random stream usage model in Vector Statistics and contains instructions on initializing, creating, saving, and restoring random streams.
Other Sections
This document has a range of other information not covered on this page. Other sections contain a typical usage model of random number generators, an overview of methods for generating random numbers from statistical distributions, and a description of the differences between the fast and accurate modes in Intel® MKL random number generators.

Product and Performance Information


Performance varies by use, configuration and other factors. Learn more at