Intel® Software Network

The Vector Math Library (VML) included with Intel® Math Kernel Library (Intel® MKL) provides highly optimized vector implementations of computationally intensive core mathematical functions. The library has both Fortran and C interfaces for all VML functions.. All functions have also been threaded (click on function to see threaded performance).


Supported VML Functions
Click the function names in the table below to see VML performance and accuracy data.


All functions are available for Real data types.
ˆ Indicates support for Complex data types.

The Vector Math Library in Intel MKL can provide substantial performance advantages over scalar implementations. The chart below compares Intel® MKL 9.1 VML functions to the equivalent functions implemented by LibM, the runtime library that supports calls to math functions from the Intel® C++ and Fortran compilers. The chart shows that even high accuracy VML functions can provide a 2-5x the performance benefit in many cases.

VML functions support Single and Double Precision and are provided with three performance/accuracy modes. Having multiple accuracy modes is a key feature that separates Intel MKL from vectorizing compilers. With Intel MKL, you can choose the precision and accuracy that best meets your needs and thus maximizes performance of your specific code.

Click here for detailed VML Performance and Accuracy tables of all functions on various Intel® processor-based systems.

Use the Vector Math Library in Intel MKL to ensure you get maximum performance instead of relying on your compiler to vectorize your code. The Intel MKL Vector Math Library gives you more control and can massively increase the performance of your application compared to non-vectorized scalar functions from compiler runtime libraries.

^§ Performance tests and ratings are measured using specific computer systems and/or components and reflect the appropriate performance of Intel products as measured by those tests. Any difference in system design or configuration may affect actual performance. Buyers should consult other sources of information to evaluate the performance of systems or components they are considering purchasing. For more information on performance tests and on the performance of Intel products, go to http://www.intel.com/software/products/.