PRODUCT-SHUFFLE NETWORKS - TOWARD RECONCILING SHUFFLES AND BUTTERFLIES

被引:34
作者
ROSENBERG, AL
机构
[1] Department of Computer and Information Science, University of Massachusetts, Amherst
基金
美国国家科学基金会;
关键词
INTERCONNECTION NETWORK; PARALLEL ARCHITECTURE; NETWORK EMULATION; DIRECT-PRODUCT NETWORK; BUTTERFLY NETWORK; CUBE-CONNECTED CYCLES NETWORK; SHUFFLE-EXCHANGE NETWORK; DEBRUIJN NETWORK;
D O I
10.1016/0166-218X(92)90152-Z
中图分类号
O29 [应用数学];
学科分类号
070104 ;
摘要
We study product-shuffle (PS) net works, which are direct products of de Bruijn networks, as interconnection networks for parallel architectures. PS networks can be viewed as generalizing both butterfly-oriented networks (such as the butterfly and cube-connected cycles networks) and shuffle-oriented networks (such as the de Bruijn and shuffle-exchange networks), in the sense that . PS networks can emulate both butterfly-oriented and shuffle-oriented networks of any size, via emulations that are work preserving, i.e., preserve the processor-time product; . PS networks share many computationally valuable structural features of various butterfly- and shuffle-oriented networks, including pancyclicity, logarithmic diameter, and large complete binary tree subnetworks; . PS networks overcome certain computational deficiencies of butterfly- and shuffle-oriented networks, by containing as subnetworks moderate-size meshes and meshes of trees, networks which butterfly- and shuffle-oriented networks cannot emulate efficiently. Finally, PS networks attain their communication power at modest cost: they are 8-valent, and they enjoy VLSI layouts that consume only modestly more area than the best layouts of like-sized butterfly- and shuffle-oriented networks.
引用
收藏
页码:465 / 488
页数:24
相关论文
共 24 条
[11]   OPTIMAL EMBEDDINGS OF BUTTERFLY-LIKE GRAPHS IN THE HYPERCUBE [J].
GREENBERG, DS ;
HEATH, LS ;
ROSENBERG, AL .
MATHEMATICAL SYSTEMS THEORY, 1990, 23 (01) :61-77
[12]  
LEIGHTON FT, 1983, COMPLEXITY ISSUES VL
[13]  
LEIGHTON FT, 1984, 1983 WORKSH GRAPH TH, P200
[14]   THE CUBE-CONNECTED CYCLES - A VERSATILE NETWORK FOR PARALLEL COMPUTATION [J].
PREPARATA, FP ;
VUILLEMIN, J .
COMMUNICATIONS OF THE ACM, 1981, 24 (05) :300-309
[15]  
Ranade A. G., 1987, 28th Annual Symposium on Foundations of Computer Science (Cat. No.87CH2471-1), P185, DOI 10.1109/SFCS.1987.32
[16]   A LOGARITHMIC TIME SORT FOR LINEAR SIZE NETWORKS [J].
REIF, JH ;
VALIANT, LG .
JOURNAL OF THE ACM, 1987, 34 (01) :60-76
[17]   TOPOLOGICAL PROPERTIES OF HYPERCUBES [J].
SAAD, Y ;
SCHULTZ, MH .
IEEE TRANSACTIONS ON COMPUTERS, 1988, 37 (07) :867-872
[18]   THE DEBRUIJN MULTIPROCESSOR NETWORK - A VERSATILE PARALLEL PROCESSING AND SORTING NETWORK FOR VLSI [J].
SAMATHAM, MR ;
PRADHAN, DK .
IEEE TRANSACTIONS ON COMPUTERS, 1989, 38 (04) :567-581
[19]  
STANFILL C, 1987, HA873 THINK MACH COR
[20]   PARALLEL PROCESSING WITH PERFECT SHUFFLE [J].
STONE, HS .
IEEE TRANSACTIONS ON COMPUTERS, 1971, C 20 (02) :153-&