Solaris Studio Topics: Computing

The floating-point environment on Oracle Sun SPARC and x86/x64 platforms enables you to develop robust, high-performance, portable numerical applications. The floating-point environment can also help investigate unusual behavior of numerical programs written by others. These systems implement the arithmetic model specified by IEEE Standard 754 for Binary Floating Point Arithmetic.

Sun Performance Library is a set of optimized, high-speed mathematical subroutines for solving linear algebra and other numerically intensive problems. Sun Performance Library is based on a collection of public domain library functions available from Netlib ( ) that have been enchanced and optimized for high-performance platforms.

Sun Performance Library contains enhanced versions of the following standard libraries:

  • LAPACK version 3.1.1 for solving linear algebra problems.
  • BLAS1 (Basic Linear Algebra Subprograms) for performing vector-vector operations.
  • BLAS2 for performing matrix-vector operations.
  • BLAS3 for performing matrix-matrix operations.
  • Netlib Sparse-BLAS for performing sparse vector operations.
  • NIST Fortran Sparse BLAS version 0.5 for performing fundamental sparse matrix operations.
  • SuperLU version 3.0 for solving sparse linear systems of equations.
  • ScaLAPACK 1.8.0 version of LAPACK for distributed memory systems using Sun ClusterTools 8.1 based on OpenMPI 1.3.

Sun Performance Library includes the following additional routines:

  • Fast Fourier transform (FFT) routines
  • Direct Sparse Solver routines 

Solaris Studio Documentation

Reference Manuals

  • Numerical Computation Guide
    A complete application programmer's handbook to understanding the data structures and operations made available by hardware, system software, and software libraries that together implement IEEE Standard 754. IEEE Standard 754 makes it easier to write numerical applications. It is a solid, well-thought-out basis for computer arithmetic that advances the art of numerical programming. (November, 2005) 
  • Sun Performance Library Reference Manual
    This manual reproduces the section 3P man pages, and provides detailed synopses of all the routines in the library. (June, 2007)
See What's New In Oracle Solaris Studio 12.3 for information on new and changed features in the current release of the math library and Sun Performance Library.

Technical Articles

Using the Tuned Libraries for Peak Application Performance 
This technical white paper discusses the Sun Performance Library (Perflib) component of Oracle Solaris Studio, and demonstrates how it can be used in applications to boost performance significantly via its highly optimized mathematical subroutines for SPARC® and x86 processor-based systems. (January 2011)

Interval Angles and the Fortran ATAN2 Function (pdf)
Issues that come up when working with interval angles or any periodic numbers, specifically how to rationalize the containment set (or cset) is for the ATAN2 Fortran intrinsic function for all possible interval arguments. (November 8, 2002)

Fast Fourier Transform (FFT) Routines
New FFT interfaces with Sun Performance Library. (November 8, 2002)

Numerical Source Code Downloads

Eight correctly-rounded double-precision transcendental functions from libm. LIBMCR is supplied as source code along with makefiles to build the library and tests to check basic functionality. Envisioned is its use as a reference base for Java Numerics projects or research into, and validation of correctly rounded transcendental functions.
You can download it now. (November 12, 2004)

Further Reading

What Every Computer Scientist Should Know About Floating-Point Arithmetic by David Goldberg
Originally published in the March, 1991 issue of Computing Surveys, Copyright 1991, Association for Computing Machinery, Inc., reprinted by permission.


Performance Analyzer and Tools Forum
Discuss the use and features of the Solaris Studio Performance Analyzer with other users.

Back to Documentation