著者
三好 甫 吉岡 義朗 池田 正幸 高村 守幸 Miyoshi Hajime Yoshioka Yoshiro Ikeda Masayuki Takamura Moriyuki
出版者
航空宇宙技術研究所
雑誌
航空宇宙技術研究所特別資料 = Special Publication of National Aerospace Laboratory SP-16 (ISSN:0289260X)
巻号頁・発行日
vol.16, pp.99-106, 1991-12

In a few years, a computer which processes CFD programs over 100 times faster than the Fujitsu VP400 and has a main memory capacity of more than 32G bytes will be required for CFD technology to play an important role in aerospace research and development. A distributed main memory parallel processor is free from the memory throughput bottleneck which prevents the implementation of shared memory parallel processors with the necessary speed. In the light of regular characteristics of CFD codes, a distributed memory parallel processor is likely to deliver the above-mentioned processing speed. Its characteristics include a physically distributed main memory which logically provides programmers with global and local memory views, processing elements with high speed RISC scalar units and high speed vector units with large capacity vector registers, and a crossbar network which interconnects a large number of processing elements. Such a processor can be suitably called the "Numerical Wind Tunnel". This paper describes the basic main memory structure, system configuration, processing element, and interconnection network and communication mechanism of the Numerical Wind Tunnel.
著者
三好 甫 Miyoshi Hajime
出版者
航空宇宙技術研究所
雑誌
航空宇宙技術研究所特別資料 = Special Publication of National Aerospace Laboratory SP-13 (ISSN:0289260X)
巻号頁・発行日
vol.13, pp.1-26, 1990-09

It is well known that the rapid development in computational fluid dynamics (CFD) has followed the development of powerful computers. In this report, we show firstly that the same situation is also recognized at NAL. Secondly, we give the requirements for computer power (processing speed and main memory capacity) that is necessary for the CFD to play important roles in the future R&D of aerospace planes and inovative aircraft,and we point out that today's supercomputer architectures are not likely to attain the processing speed required and that more parallelism should be developed. Finally, we discuss several important items in designing the target computer, bipolar technologybased processing element vs. MOS technology based processing element, shared memory vs. distributed memory, interconnection network's topology, software technologies' cost and running cost of the target computer and so on.
著者
三好 甫 Miyoshi Hajime
出版者
航空宇宙技術研究所
雑誌
航空宇宙技術研究所特別資料 = Special Publication of National Aerospace Laboratory SP-16 (ISSN:0289260X)
巻号頁・発行日
vol.16, pp.91-98, 1991-12

In this paper, it is shown that the Numerical Wind Tunnel (NWT) with an actual performance of more than 100 times higher than Fujitsu-VP400 is feasible using a multicomputer architecture and crossbar interconnection network.
著者
三好 甫 MIYOSHI Hajime
出版者
航空宇宙技術研究所
雑誌
航空宇宙技術研究所報告 = Technical Report of National Aerospace Laboratory TR-1108 (ISSN:03894010)
巻号頁・発行日
vol.1108, pp.29, 1991-05

計算空気力学(CFD)の研究を推進し,その成果を航空機および航空宇宙機の研究開発に役立たせる為には,CFDプログラムの処理能力がFujitsu VP400の100倍以上のCFD専用の計算機-超高速数値風洞(UHSNWT)をここ2~3年のうちに,更にこれを10倍以上上廻るUHSNWTを1990年代末に開発することが是非とも必要となる。報告はこのUHSNWTが備えるべき要件について,開発・運転コスト,信頼性,CFD計算法に対する適合性,運用環境等の視点から論ずる。同時に報告はこれ等の要件を満足するUHSNWTの構想を 1.これ迄のベクトル計算機の使用経験 2.種々のベクトル計算機モデルのベクトル処理に対するソフトウェアシミュレーション 3.LSI技術の将来の発展動向 に基づいて述べる。報告はVP400の100倍以上の性能のUHSNWTはクロスバーネットワークを採用した分散主記憶型並列計算機により実現できることを結論付けている。