The PECM group
supports a variety of mathematical software, libraries and tools on NCSA's HPC
resources.
A current list is available from the
HPC
Software Repository, specifically the
Mathematics and Statistics and
Numerical Programs and Routines entries.
Here we only highlight some of the software and tools available.
Math Software
NCSA supports open-source and commercial software on its HPC resources.
BLAS and LAPACK
Basic Linear Algegra Subprogram (BLAS) and Linear Algebra PACKage Libraries (LAPACK)
provide the basis of a substantial number of scientific codes. For a performance comparison of a
few of the more common BLAS routines from the various sources, see the
BLAS Comparison page.
- ESSL - IBM's Scientific Subprogram Library.
- MKL - Intel's Math Kernel Library for BLAS, LAPACK and more.
- Goto - High Performance BLAS libraries.
- ATLAS - Open-source BLAS and some LAPACK.
ScaLAPACK
ScaLAPACK provides a parallel implementation of BLAS and LAPACK. It is available from a open-source
distribution as well as from IBM's PESSL.
Fourier Transforms
For a performance comparison of FFT packages available, see the FFT Comparison page.
- ESSL - IBM's Scientific Subprogram Library
- MKL - Intel's Math Kernel Library
- FFTW - The Fastest Fourier Transform in the West
PETSc
PETSc
(Portable, Extensible Toolkit for Scientific Computation) is a suite of data structures and
routines for the uni- and parallel-processor solution of large-scale scientific application problems
modeled by partial differential equations.
SPRNG
Parallel random number generation is provided by the
SPRNG
library. Originally designed for parallel Monte Carlo applications, it provides a method for rapid
generation of a larger amount of random numbers with no correlation.
Tools
Totalview
The parallel debugger Totalview
provides a convenient method for debugging parallel applications on clusters and SMP machines.
Contact:
Contact mathsoft@ncsa.uiuc.edu for more
information on libraries that are available or that you think
should be available on NCSA HPC machines.