MathKeisan 1.6.0 for SX Release Notes


January 2005
NEC Corporation

Introduction

MathKeisan is a highly-tuned collection of mathematical libraries for NEC high performance computers. There is a version of MathKeisan for the NEC SX vector machines running SUPER-UX, and a version for NEC Itanium® Processor Family (IPF) servers running Linux. Unless otherwise noted, all references to MathKeisan in these Release Notes are to MathKeisan for SUPER-UX.

MathKeisan contains assembly code for BLAS subprograms. There are shared memory parallel versions of BLAS, FFT, SOLVER and LAPACK. The BLACS (called by ScaLAPACK) and ParMETIS libraries are distributed memory parallel, using MPI.

The libraries in MathKeisan are listed in Table 1.

Table 1: Libraries in MathKeisan
Name Description
BLAS Basic Linear Algebra Subprograms
LAPACK Linear Algebra PACKage for high performance computers
ScaLAPACK Scalable Linear Algebra PACKage (contains PBLAS)
BLACS Basic Linear Algebra Communication Subprograms (uses MPI)
PARBLAS Shared memory parallel BLAS
CBLAS C interface to BLAS
SBLAS Sparse BLAS (from ACM Algorithm 692)
FFT Fast Fourier Transforms with HP VECLIB interface and CRAY LIBSCI 3.1 interface
PARFFT Shared memory parallel FFT with HP VECLIB interface and CRAY LIBSCI 3.1 interface
METIS Matrix/Graph ordering and partitioning library
ParMETIS Parallel Matrix/Graph ordering and partition library (uses MPI)
SOLVER Direct solver for sparse symmetric systems
ARPACK Solution of large scale eigenvalue problems

Compatibility

For a list of machines and SUPER-UX revisions compatible with MathKeisan 1.6.0, please follow the "Compatibility" link at www.mathkeisan.com. The following compilers and MPI were used to build the MathKeisan libraries:

f90 for SX, Rev.305 C++/SX, Rev.064 MPI/SX: library Version 6.7.18

New in MathKeisan 1.6.0 for SX

Installation

The MathKeisan 1.6.0 distribution for SX contains these files:

The README file has information on how to install MathKeisan 1.6.0.

User's Guide

For more information, see the MathKeisan User's Guide (also available at www.mathkeisan.com). The User's Guide is organized as follows:

  1. Acknowledgments
  2. Overview of Libraries
  3. Documentation
  4. Linking with MathKeisan
  5. User Notes
  6. Routine List