[Projects for eigenvalue computation]
Project ABC-LIB:
This project is an ongoing project of

the organization head.

This project focused on developing the high performance

numerical library in Japan, especially parallel algorithms

and implementations for PC-Clusters.

http://www.abc-lib.org/

Project HINTS:
The organization head was a member of this project.

This project was one of the early projects for

developing high performance parallel numerical library

in Japan.

http://www.hints.org/
The Oxford Eigenvalue Project:
This is a research project of The university of OXFORD,

Computing Laboratory. The project is focused on new method

for very ill conditioned problems. Numerical analysis,

numerical computation algorithms and software, and application

of industry are also the aims of this project.

http://web.comlab.ox.ac.uk/oucl/research/grants/aw.html
A Study on parallel eigenvalue computation algorithm:
This project partially supported on Fuji Research Institute

Corporation.

http://www.fuji-ric.co.jp/ccse/project/ngcse/parallel/eigen/
Project PRISM:
The PRISM (Parallel Research on Invariant Subspace Methods) project

focuses on developing basics of parallel eigenvalue computation

and algorithms.

Presently, the aim of this study is the Invariant Subspace

Decomposition algorithm for dense symmetric matrices.

http://www-unix.mcs.anl.gov/prism/
[Library for eigenvalue computation]
ILIB:
The head was also a core member for developing this library.

ILIB is the first library, which contains auto-tuning facilities

for parallel numerical library. The object codes for Japan super-computer

are only opened in this project (Presently, the source codes is not opened.)

http://www.super-computing.org/~katagiri/nadia.html
LAPACK:
This library is one of the most famous libraries for linear algebra
library.

Several kinds of solvers, for example eigenvalue problem and linear
equation

solvers, contain this library. In addition, the source codes of them are
free

and opened
.
http://www.netlib.org/lapack/
ScaLAPACK:
This is a parallel library for LAPACK. This library has a high portability

based on LAPACK.


http://www.netlib.org/scalapack/
Scilab:
This is one of numerical libraries for scientific and technology use.

This library was developed in FRENCH NATIONAL INSTITUTE FOR RESEARCH

IN COMPUTER SCIENCE AND CONTROL(INRIA).

http://www-rocq.inria.fr/scilab/
PARCEL:
This parallel numerical library was developed in Japan Atomic Energy

Research Institute, Center for Promotion of Computational Science and

Engineering.

http://guide.tokai.jaeri.go.jp/program/
ARPACK:
This is an eigensolver using Implicitly Restarted Arnoldi method.

http://www.caam.rice.edu/software/ARPACK/
[References]
Area4 Working Notes:
This reference is the result of joint project of Fujitsu and

Australian National University (ANU).

http://anusf.anu.edu.au/Area4_Working_Notes/
Ken Stanley’s Symmetric Eigensolver Page:
This is a reference for symmetric eigensolver by Dr. Ken Stanley.

http://http.cs.berkeley.edu/~stanley/symeig/index.html