Current State of the Art of Mathematical Software for Linear Algebra

7/21/97


Click here to start


Table of Contents

Current State of the Art of Mathematical Software for Linear Algebra

Outline

Important Sources of Software and Information

Freely Available LA SW on the Web

Directions

The Importance of Standards - Software

Why Higher Level BLAS?

LAPACK Linear Algebra Library in F77

LAPACK Ongoing Work

Challenges in Developing Distributed Memory Libaries

Heterogeneous Computing

Standards - The MPI Forum

Portability of Numerical Software

ScaLAPACK

ScaLAPACK Structure

Choosing a Data Distribution

Parallelism in ScaLAPACK

ScaLAPACK - Whatís Included

Direct Sparse Solvers

Parallel Iterative Solvers

Parallel Sparse Eigenvalue Solvers

Java

Java for Numerical Computations?

LAPACK to JAVA

Parameterized Libraries

Motivation for Network Enabled Solvers

PPT Slide

On Mathematical Software

Freely Available LA SW on the Web

References

Author: Jack Dongarra

Email: dongarra@cs.utk.edu

Home Page: http://www.netlib.org/utk/people/JackDongarra/