NAG Logo
Numerical Algorithms Group
graphic

The NAG SMP Library

Unrivalled functionality, increased speed and heightened performance for shared memory parallelism

The NAG SMP Library, used by some of the world's most prestigious super computing centers was produced to enable developers and programmers to make optimal use of the processing power and shared memory parallelism of the Symmetric Multi-Processor (SMP) systems.

A proportion of the routines within the NAG SMP Library has been specially developed and tuned to provide the utmost performance on SMP platforms. These tuned routines deliver levels of performance and scalability superior to many other products currently available. In fact, NAG has developed and pioneered many parallelized computational algorithms, often unique to NAG.

During recent benchmark studies the SMP Library, Mark 21 outperformed many comparable products often showing vast performance improvements, crucial for programmers in today's highly competitive high performance computing environment.

SMP Library Contents

Below is a list of some of the main numerical and statistical capabilities of the Library. Further details on the contents of the Library can be read here.

Key benefits of NAG SMP Library:

  • Contains over 120 specialised routines tuned for maximum performance
  • Delivers improved performance in over 250 other routines including key numerical areas such as optimisation, statistics and partial differential equations (PDEs)
  • Easy to use and link to due to identical interface to NAG Fortran Library
  • Highly flexible - assists rapid migration from serial code, and between platforms
  • Interoperable - routines can be called from multiple computing languages
  • Based on the OpenMP Program Interface (API), which supports multi-platform shared-memory parallel programming in C/C++ and Fortran on all architectures including Unix and Windows platforms.

Specialized functionality of SMP Library

  • Fast fourier transforms (FFTs)
  • Dense linear algebra (LAPACK)
  • Sparse iterative solvers
  • Sparse direct solvers (based on SuperLU)
  • Sparse iterative eigensolvers (based on ARPACK)

Other areas that benefit from tuned routines

  • Ordinary differential equations (ODEs)
  • Optimization
  • Multivariate statistics
  • Linear algebra

Statistical facilities

  • Random number generation
  • Simple calculations on statistical data
  • Correlation and regression analysis
  • Multivariate methods
  • Analysis of variance and contingency table analysis
  • Time series analysis
  • Nonparametric statistics

Replacing NAG Fortran Library routines with their SMP equivalents

The NAG SMP Library is based on, and includes, the full functionality of the NAG Fortran Library so the arguments list and routines names are identical. This provides an easy upgrade path because the same code can call a routine that performs the same function, in either the Fortran Library or the SMP Library. The distinction between the two is only made when the final code is linked to a library. Application that reference a NAG Library routine can therefore take advantage of the performance improvement simply by recompiling and linking your code on an SMP system.

No specialist parallel programming knowledge is required to apply the SMP techniques to your code. The parallelism is carried out automatically and transparently in certain distinct regions within the routines.

Online Documentation

The software in the library is accompanied by first-class online documentation giving you the detailed information you need to carry out your work quickly and effectively.

Regular Updates

To ensure that we fulfill your existing and future requirements, the library is regularly updated with new and enhanced algorithms for use in evolving business areas.

Expert Support

If you subscribe to our dedicated Customer Support Service, not only will you automatically receive any update that is produced, but you will have direct access via the NAG Response Centre to our experts, who are there to assist you with your technical queries or difficulties.

Quality Assurance

The validity of each component is tested on each of the machine ranges for which the product is available. Only when an implementation satisfies our stringent accuracy standards is it released. As a result, you can rely on the proven accuracy and reliability of the components to give you the right answers.

© The Numerical Algorithms Group 2008
Privacy Policy | Trademarks

© Numerical Algorithms Group

Visit NAG on the web at:

www.nag.co.uk (Europe and ROW)
www.nag.com (North America)
www.nag-j.co.jp (Japan)

http://www.nag.com/numeric/fl/FSdescription.asp