SQUARE MESHES ARE NOT ALWAYS OPTIMAL

被引:34
作者
BARNOY, A [1 ]
PELEG, D [1 ]
机构
[1] WEIZMANN INST SCI,DEPT APPL MATH,IL-76100 REHOVOT,ISRAEL
关键词
BROADCASTING; MESH-CONNECTED COMPUTER; MULTIPLE BUSES; PARALLEL PROCESSING; SEMIGROUP COMPUTATION;
D O I
10.1109/12.73589
中图分类号
TP3 [计算技术、计算机技术];
学科分类号
0812 ;
摘要
We consider mesh-connected computers with multiple buses, providing broadcast facilities along rows and columns. A tight bound of THETA(n1/8) is established for the number of rounds required for semigroup computations on n values distributed on a two-dimensional rectangular mesh of size n with a bus on every row and column. The upper bound is obtained for a skewed rectangular mesh of dimensions n3/8 x n5/8. This result should be compared to the tight bound of THETA(n1/6) for the same problem on the square (n1/2 x n1/2) mesh. This implies that in the presence of multiple buses, a skewed configuration may perform better than a square configuration for certain computational tasks. Our result can be extended to the d-dimensional mesh, giving a lower bound of OMEGA(n1/d2d) and an upper bound of O(d2d+1n1/d2d); these bounds are optimal within constant factors for any constant d. (Note that for d > 3, our results are mostly of theoretical interest.)
引用
收藏
页码:196 / 204
页数:9
相关论文
共 18 条
[1]  
AGGARWAL A, 1986, IEEE T COMPUT, V35, P62, DOI 10.1109/TC.1986.1676658
[2]  
BOKHARI SH, 1984, IEEE T COMPUT, V33, P133, DOI 10.1109/TC.1984.1676405
[3]   REAL-TIME COMPUTATION BY N-DIMENSIONAL ITERATIVE ARRAYS OF FINITE-STATE MACHINES [J].
COLE, SN .
IEEE TRANSACTIONS ON COMPUTERS, 1969, C 18 (04) :349-&
[4]  
CORDELLA LP, 1978, IEEE T COMPUT, V27, P904, DOI 10.1109/TC.1978.1674969
[5]   SOME COMPLEXITY RESULTS FOR MATRIX COMPUTATIONS ON PARALLEL PROCESSORS [J].
GENTLEMAN, WM .
JOURNAL OF THE ACM, 1978, 25 (01) :112-115
[6]   MULTI-MICROPROCESSOR SYSTEM FOR FINITE-ELEMENT STRUCTURAL-ANALYSIS [J].
JORDAN, HF ;
SAWYER, PL .
COMPUTERS & STRUCTURES, 1979, 10 (1-2) :21-29
[7]   CELLULAR INTERCONNECTION ARRAYS [J].
KAUTZ, WH ;
LEVITT, KN ;
WAKSMAN, A .
IEEE TRANSACTIONS ON COMPUTERS, 1968, C 17 (05) :443-&
[8]  
KOSARAJU SR, 1979, 11 ANN ACM S THEOR C, P231
[9]   PARALLEL PICTURE PROCESSING MACHINE [J].
KRUSE, B .
IEEE TRANSACTIONS ON COMPUTERS, 1973, C 22 (12) :1075-1087
[10]   ARRAY PROCESSOR WITH MULTIPLE BROADCASTING [J].
KUMAR, VKP ;
RAGHAVENDRA, CS .
JOURNAL OF PARALLEL AND DISTRIBUTED COMPUTING, 1987, 4 (02) :173-190