Watson Sparse Matrix Package (WSMP) is a collection of algorithms for efficiently solving large systems of linear equations whose coefficient matrices are sparse. This high-performance, robust, and easy-to-use software can be used as a serial package, or in a shared-memory multiprocessor environment, or as a scalable parallel solver in a message-passing environment, where each node can either be a uniprocessor or a shared-memory multiprocessor.

A sparse matrix is one that has relatively few non-zero (or "interesting") entries. Like a large spreadsheet with very few entries, a sparse matrix can represent challenges to memory efficiency. Sophisticated algorithms can store and manipulate the information in such a matrix requiring much less space than a dense or full matrix of the same size.

How does it work?

While most WSMP users apply it to solve sparse linear systems resulting from partial differential equations (PDEs) in finite-difference, finite-volume, and finite-element applications, WSMP has built-in support for both barrier (interior-point) and simplex methods for solving linear programming (LP) problems. On a wide range of sparse linear systems and hardware platforms, WSMP has been shown to be significantly faster than other similar software.

Inserting calls to WSMP routines for solving the linear systems (instead of writing one's own code or using some other software) could significantly reduce the execution time of the application. This applies to linear systems arising in the solution of linear programming problems for optimization or from partial differential equations in scientific and engineering applications, such as circuit and device simulation or sheet-metal forming.

Anshul Gupta received a B.Tech. degree from the Indian Institute of Technology, New Delhi, in 1988 and a Ph.D. from the University of Minnesota in 1995, both in Computer Science. He is currently a research staff member at the IBM® T. J. Watson Research Center, Yorktown Heights, N.Y. Dr. Gupta's research interests include parallel algorithms, sparse matrix computations, and applications of parallel processing in scientific computing and optimization.