Introduction

Abstract

The remarkable progress in theoretical condensed-matter physics depends on the development and improvement of the methodology and also on the rapid progress in modern high-performance computers. As for the latter, although the performance of a single processor is still being improved, parallel computing will become increasingly important. Williams discussed the parallel computing from various aspects. It may be interesting to cite some of the subsection titles from his article: “Improved Pain/Gain Ratio”, “Cost/Performance Versus Performance”, “Budget-Limited Versus Time-Limited Computing”, “Parallelism will be Medium-Grained”, “Shared Versus Distributed Memory”. A unique idea about the order-N method in the electronic-structure calculation was also presented and its relation to parallel computing was discussed.