Vector performance prediction of kernel loops on Earth Simulator

Yokokawa, Mitsuo; Saito, Minoru*; Hagiwara, Takashi*; Isobe, Yoko*; Jinguji, Satoshi*

Nihon Keisan Kogakkai Rombunshu, 4, p.31 - 36, 2002/00

http://joi.jlc.jst.go.jp/JST.JSTAGE/jsces/2001.20010040

Earth simulator is a distributed memory parallel system which consists of 640 processor nodes connected by a full crossbar network. Each processor node is a shared memory system which is composed of eight vector processors. The total peak performance and main memory capacity are 40Tflops and 10TB, respectively. A performance prediction system GS for the Earth Simulator has been developed to estimate sustained performance of programs. To validate accuracy of vector performance prediction by the GS, the processing times for three groups of kernel loops estimated by the GS are compared with the ones measured on SX-4. It is found that the absolute relative errors of the processing time are 0.89%,1.42% and 6.81% in average for three groups. The sustained performance of three groups on a processor of the Earth Simulator have been estimated by the GS and those performance are 5.94Gflops,3.76Gflops and 2.17Gflops in average.

Practice in multi-disciplinary computing; Transonic aero-structural dynamics of semi-monocoque wing

Onishi, Ryoichi*; Guo, Z.*; Kimura, Toshiya*; Iwamiya, Toshiyuki*

Proceedings of 4th International Conference on Supercomputing in Nuclear Applications (SNA 2000) (CD-ROM), 12 Pages, 2000/09

no abstracts in English

TME (Task Mapping Editor): Tool for excuting distributed Parallel Computing; TME user's manual

Takemiya, Hiroshi*; Yamagishi, Nobuhiro*; Imamura, Toshiyuki; Ueno, Koichi*; Koide, Hiroshi; Tsujita, Yuichi; Hasegawa, Yukihiro*; Higuchi, Kenji; Matsuda, Katsuyuki*; Hirayama, Toshio

JAERI-Data/Code 2000-013, p.52 - 0, 2000/03

JAERI-Data-Code-2000-013.pdf:8.64MB

no abstracts in English

TME (Task Mapping Editor): Tool for executing distributed parallel computing; Design report on TME

Takemiya, Hiroshi*; Yamagishi, Nobuhiro*; Imamura, Toshiyuki; Ueno, Koichi*; Koide, Hiroshi; Tsujita, Yuichi; Hasegawa, Yukihiro*; Higuchi, Kenji; Matsuda, Katsuyuki*; Hirayama, Toshio

JAERI-Data/Code 2000-010, p.49 - 0, 2000/02

JAERI-Data-Code-2000-010.pdf:2.04MB

no abstracts in English

Multidisciplinary aero-structural modeling on parallel computers

Onishi, Ryoichi*; ; Guo, Z.*; *

CEAS/AIAA/ICASE/NASA Langley Int. Forum on Aeroelasticity and Strucrual Dynamics 1999, (2), p.483 - 489, 1999/06

no abstracts in English

Analytical model for aero-structural interaction problem of a wing-box structure

Onishi, Ryoichi*; ; Guo, Z.*; *

A Collection of the 17th AIAA Applied Aerodynamics Conf. Technical Papers, p.492 - 496, 1999/00

no abstracts in English

Parallel processing for the direct simulation Monte Carlo method

Yokokawa, Mitsuo; Watanabe, Kenji*; *; *;

Computational Fluid Dynamics Journal, 1(3), p.337 - 346, 1992/10

no abstracts in English

Improvement of parallel visualization efficiency by pre-processing large-scale distributed data

Guo, Z.; Nishida, Akemi; Choi, B.; Nakajima, Norihiro

no journal, , 

In the field of seismic analysis of nuclear facilities, large-scale parallel analyses using numerical models with several hundred millions of DOFs are becoming possible by the recent advances in high-performance parallel computing technologies. In dealing with such three dimensional time series data, the post-processing may be often more difficult than the seismic response simulation itself. The purpose of the current study is to develop a parallel visualization application, which can visualize large-scale simulation results (distributed time series data) effectively. In this report, we show an approach to increase the efficiency of parallel visualization by more than 200 times by using appropriate pre-processing for this kind of large-scale distributed time series data.