Performance evaluation of massively parallel processing architectures with three-dimensional optical interconnections

被引:9
作者
Betzos, GA [1 ]
Mitkas, PA
机构
[1] Colorado State Univ, Ctr Optoelect Comp Syst, Ft Collins, CO 80523 USA
[2] Colorado State Univ, Dept Elect Engn, Ft Collins, CO 80523 USA
来源
APPLIED OPTICS | 1998年 / 37卷 / 02期
关键词
D O I
10.1364/AO.37.000315
中图分类号
O43 [光学];
学科分类号
070207 ; 0803 ;
摘要
We evaluate the performance of three-dimensional optoelectronic computer architectures on the basis of basic database operations and parallel benchmark algorithms for numerical computations. We show that the select and the join database operations can be performed much faster with an optical interconnection network. Also, optoelectronic architectures can pet-form the fast Fourier transform and sorting benchmarks orders of magnitude faster than electronic supercomputers. An architecture with an adequately fast reconfigurable interconnection network can perform the conjugate-gradient benchmark faster than all parallel supercomputers, but its performance is not as impressive when a fixed network is used. In the case of the multigrid benchmark the three-dimensional optoelectronic architecture also can outperform the best parallel supercomputers. (C) 1998 Optical Society of America.
引用
收藏
页码:315 / 325
页数:11
相关论文
共 15 条
[11]   AN ADAPTATION OF FAST FOURIER TRANSFORM FOR PARALLEL PROCESSING [J].
PEASE, MC .
JOURNAL OF THE ACM, 1968, 15 (02) :252-+
[12]  
SAINI S, 1996, NAS9618 NASA AM RES
[13]  
SCHENFELD E, 1995, 1995 OSA, V10, P16
[14]   FOLDED PERFECT SHUFFLE OPTICAL PROCESSOR [J].
STIRK, CW ;
ATHALE, RA ;
HANEY, MW .
APPLIED OPTICS, 1988, 27 (02) :202-203
[15]   PARALLEL PROCESSING WITH PERFECT SHUFFLE [J].
STONE, HS .
IEEE TRANSACTIONS ON COMPUTERS, 1971, C 20 (02) :153-&