Square Meshes are not always Optimal

Mesh-connected computers with multiple buses providing broadcast facilities along rows and columns are discussed. A tight bound of Theta (n/sup 1/8/) 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 n/sup 3/8/*n/sup 5/8/. This result is compared to the tight bound of Theta (n/sup 1/6/) for the same problem on the square (n/sup 1/2/*n/sup 1/2/) mesh. It is shown that in the presence of multiple buses, a skewed configuration may perform better than a square configuration for certain computational tasks. The result can be extended to the d-dimensional mesh, giving a lower bound of Omega (n/sup 1/d alpha /) and an upper bound of O(d2/sup d+1/ n/sup 1/d alpha /), where alpha =2/sup d/; these bounds are optimal within constant factors for any constant d. It is noted that for d>3, the results of are mostly of theoretical interest. >

[1]  Harry F. Jordan,et al.  A multi-microprocessor system for finite element structural analysis , 1979 .

[2]  W. Morven Gentleman,et al.  Some Complexity Results for Matrix Computations on Parallel Processors , 1978, JACM.

[3]  S. Rao Kosaraju,et al.  Fast parallel processing array algorithms for some graph problems(Preliminary Version) , 1979, STOC.

[4]  Quentin F. Stout,et al.  Mesh-Connected Computers with Broadcasting , 1983, IEEE Transactions on Computers.

[5]  S. H. Unger,et al.  A Computer Oriented toward Spatial Problems , 1899, Proceedings of the IRE.

[6]  Björn Kruse,et al.  A Parallel Picture Processing Machine , 1973, IEEE Transactions on Computers.

[7]  Quentin F. Stout,et al.  BROADCASTING IN MESH-CONNECTED COMPUTERS. , 1982 .

[8]  Shahid H. Bokhari,et al.  Finding Maximum on an Array Processor with a Global Bus , 1984, IEEE Transactions on Computers.

[9]  Stephen N. Cole Real-Time Computation by n-Dimensional Iterative Arrays of Finite-State Machines , 1969, IEEE Trans. Computers.

[10]  Luigi P. Cordella,et al.  An Analysis of Computational Cost in Image Processing: A Case Study , 1978, IEEE Transactions on Computers.

[11]  Cauligi S. Raghavendra,et al.  HMESH: A VLSI Architecture for Parallel Processing , 1986, CONPAR.

[12]  Anthony P. Reeves,et al.  A Systematically Designed Binary Array Processor , 1980, IEEE Transactions on Computers.

[13]  H. T. Kung,et al.  Sorting on a mesh-connected parallel computer , 1977, CACM.

[14]  Quentin F. Stout,et al.  Meshes with multiple buses , 1986, 27th Annual Symposium on Foundations of Computer Science (sfcs 1986).

[15]  Karl N. Levitt,et al.  Cellular Interconnection Arrays , 1968, IEEE Transactions on Computers.

[16]  Alok Aggarwal Optimal Bounds for Finding Maximum on Array of Processors with k Global Buses , 1986, IEEE Transactions on Computers.