Professor Haruo Niimi
|Area and Subject Taught||Computer Architecture|
|Research Theme(s)||Architectures of Parallel Computers and Parallel Computing|
|Academic Degrees||Doctor of Engineering, Kyoto University|
|Keywords for Research Field||Parallel Processing, Parallel Programming,Computer Architecture|
|Office Phone Number||Not Public|
Techniques for parallel processing are absolutely essential for achieving dramatic improvements in computer system performance. Therefore, it is necessary to conduct a broad range of research, covering everything from hardware architecture to operating systems, programming languages and applications, in order to establish techniques for integrated parallel computer systems. In the last few years, we have mainly been focused on "cluster systems," consisting of multiple personal computers (PCs) or workstations (WSs) interconnected via a network. We have built such systems, evaluated their performance, developed parallel programs, and conducted research on support systems. By designing and developing actual systems, and comparing them with systems using different parallel processing methods such as shared-memory multiprocessors, we qualitatively and quantitatively examine the various tradeoffs in terms of hardware and software for achieving efficient parallel processing systems.
Notable Publications and Works in the Last Three Years
- Tomita, Fujii ed.：情報社会とコンピュータ，「第8章 データ表現」分担執筆，pp.81-97，昭晃堂，2005.
- Ueba, Niimi：OpenMP 型並列プログラミングモデルのクラスタ計算環境への適用,京都産業大学先端科学技術研究所所報, 3, pp.41-68, 2004.
- A. Nunome, H. Hirata, H. Niimi, K. Shibayama : Performance Evaluation of Dynamic Load Balancing Scheme with Load Prediction Mechanism Using the Load Growing Acceleration for Massively Parallel Computers, Systems and Computers in Japan (Wiley), 35 (11), pp.69-79, 2004.