This is the top of the page.
Jump to main content.

Please note that JavaScript and style sheet are used in this website,
Due to unadaptability of the style sheet with the browser used in your computer, pages may not look as original.
Even in such a case, however, the contents can be used safely.

Orchestrating a brigher world


  1. Home
  2. User's Guide
  3. User's Guide: Introduction

User's Guide: Introduction

MathKeisan is a tuned collection of mathematical libraries for NEC high-performance computers. There is a version for the NEC SX vector machines running SUPER-UX.

MathKeisan contains assembly code for BLAS subprograms. There are shared memory parallel (OpenMP) versions of BLAS, FFT. There are distributed memory parallel (MPI) libraries: BLACS, ScaLAPACK and PARPACK.

MathKeisan libraries are listed in Table 1.

Table 1: Libraries in MathKeisan.
Name Description
ARPACK Solution of large scale eigenvalue problems
BLACS Basic Linear Algebra Communication Subprograms (uses MPI)
BLAS Basic Linear Algebra Subprograms
CBLAS C interface to BLAS
FFT Fast Fourier Transforms
LAPACK Linear Algebra PACKage for high performance computers
PARBLAS Shared memory parallel BLAS
PARFFT Shared memory parallel FFT
PARPACK Parallel ARPACK (uses MPI)
SBLAS Sparse BLAS (from ACM Algorithm 692)
ScaLAPACK Scalable Linear Algebra PACKage (contains PBLAS)

Document Conventions

Various fonts are used in the text, depending on what the text is for:

  • Plain text.
  • Dummy variables such as xx, or mathematical formulas such as x2+y.
  • Commands or code that the user may type in.
  • Output that the program prints out.
  • Pathnames like /opt/MathKeisan, environment variables like OMP_NUM_THREADS, library names like libblas.a, and Fortran/C code which is descriptive, such as real(kind=4).
Copyright © NEC MathKeisan® 2019. All rights reserved.