This Article 
   
 Share 
   
 Bibliographic References 
   
 Add to: 
 
Digg
Furl
Spurl
Blink
Simpy
Google
Del.icio.us
Y!MyWeb
 
 Search 
   
The Fat-Pyramid and Universal Parallel Computation Independent of Wire Delay
December 1994 (vol. 43 no. 12)
pp. 1358-1364

This paper shows that a fat-pyramid of area /spl Theta/(A) requires only O(log A) slowdown to simulate any competing network of area A under very general conditions. The result holds regardless of the processor size (amount of attached memory) and number of processors in the competing networks as long as the limitation on total area is met. Furthermore, the result is valid regardless of the relationship between wire length and wire delay. We especially focus on elimination of the common simplifying assumption that unit time suffices to traverse a wire regardless of its length, since the assumption becomes more and more untenable as the size of parallel systems increases. This paper concentrates on simulation using transmission lines (wires along which bits can be pipelined) with the message routing schedule set up off line, but it also discusses the extension to on-line simulation. This paper also examines the capabilities of a fat-pyramid when matched against a substantially larger network and points out the surprising difficulty of doing such a comparison without the unit wire delay assumption.

[1] A. V. Aho, J. E. Hopcroft, and J. D. Ullman,The Design and Analysis of Computer Algorithms. Menlo Park, CA: Addison-Wesley, 1974.
[2] R. Aleliunas and A. L. Rosenberg, "On embedding rectangular grids in square grids,"IEEE Trans. Comput., vol. C-31, no. 9, pp. 907-913, Sept. 1982.
[3] P. Bay and G. Bilardi, "Deterministic on-line routing on area-universal networks," in31st Annu. Symp. Foundations of Comput. Sci., IEEE Computer Society Press, 1990, pp. 297-306.
[4] P. E. Bay, "Area-universal interconnection networks for VLSI parallel computers," Ph.D. thesis, Dep. of Comput. Sci., Cornell Univ., May 1992.
[5] J. L. Bentley, D. Haken, and J. B. Saxe, "A general method for solving divide-and-conquer recurrences." Tech. Reu. CMU-CS-78-154. Dep. of Comput. Sci., Carnegie-Mellon Univ., Dec. 1978.
[6] S. N. Bhatt and F. T. Leighton, "A framework for solving VLSI graph layout problems,"J. Comput. Syst. Sci., vol. 28, no. 2, pp. 300-343, Apr. 1984.
[7] R. P. Brent and H. T. Kung, "Fast algorithms for manipulating formal power series,"J. ACM, vol. 25, no. 4, pp. 581-595, Oct. 1978.
[8] A. M. Despain and D. A. Patterson, "X-tree: A tree structured multiprocessor computer architecture," inProc. Fifth Int. Symp. Comput. Architecture, Apr. 1978, pp. 144-151.Comput. Architecture, pp. 21-28, Dec. 1973.
[9] R. I. Greenberg, "Efficient interconnection schemes for VLSI and parallel computation," Ph.D. thesis, Dep. of Elect. Eng. and Comput. Sci., Massachusetts Inst. of Technol., Aug. 1989, MIT/LCS/TR-456.
[10] R. I. Greenberg,"The fat-pyramid: A robust network for parallel computation," inAdvanced Research in VLSI: Proceedings of the Six MIT Conference, W. J. Dally, Ed. Cambridge, MA: MIT Press, Apr. 1990, pp. 195-213.
[11] R. I. Greenberg and C. E. Leiserson, "A compact layout for the three-dimensional tree of meshes,"Appl. Math. Lett., vol. 1, no. 2, pp. 171-176, 1988 (Also see erratum in vol. 1, no. 3, p. 315).
[12] R. I. Greenberg and C. E. Leiserson, "Randomized routing on fat-trees," inAdvances in Computing Research (Randomness and Computation, vol. 5) S. Micali, Ed. Greenwich, CT: JAI Press, 1989, pp. 345-374.
[13] R. I. Greenberg and H.-C. Oh, "Packet routing in networks with long wires, " inProc. 30th Annu. Allerton Conf. Commun., Contr. and Computing, pp. 664-673, 1992, (Revised versions: Univ. of Maryland Tech. Rep. UMIACS-TR-93-22;J. Parallel Distrib. Computing, to appear).
[14] F. T. Leighton, B. M. Maggs, A. G. Ranade and S. B. Rao, "Randomized routing and sorting on fixed-connection networks,"J. Algorithms, vol. 17, no. 1, pp. 157-205, July 1994.
[15] F. T. Leighton, B. M. Maggs, and S. B. Rao, "Packet routing and job-shop scheduling inO(congestion + dilation) steps,"Combinatorica, vol. 14, no. 2, pp. 167-180, 1994.
[16] T. Leighton, B. Maggs, and S. Rao, "Universal packet routing algorithms,"Proc. 29th IEEE Symp. Foundations Comput. Sci., 1988, pp. 256-269.
[17] C. E. Leiserson, "Area-efficient graph layouts (for VLSI)," in21st Annu. Symp. on Foundat. Comput. Sci., IEEE Computer Society Press, 1980, pp. 270-281.
[18] C. E. Leiserson, "Fat-trees: Universal networks for hardware-efficient supercomputing,"IEEE Trans. Comput., vol. C-34, pp. 892-900, Oct. 1985.
[19] C. E. Leiserson, "VLSI theory and parallel supercomputing," inAdvanced Research in VLSI: Proceedings of the Decennial Caltech Conference on VLSI, C. L. Seitz, Ed. Cambridge, MA: MIT Press, 1989, pp. 5-16.
[20] C. E. Leiserson, "The network architecture of the connection machine CM-5,"Proc. ACM Symp. Parallel Algorithms and Architectures, July 1992.
[21] C. E. Leiserson and B. M. Maggs, "Communication-efficient parallel algorithms for distributed random-access machines,"Algorithmica, vol. 3, pp. 53-77, 1988.
[22] S. L. Scott and J. R. Goodman. "The impact of pipelined channels onk-aryn-cube networks,"IEEE Trans. Parallel Disrrib. Syst., vol. 5, no. 1, pp. 2-16, Jan. 1994.
[23] C. H. Séquin, A. M. Despain, and D. A. Patterson, "Communication in X-TREE, a modular multiprocessor system," inACM 78: Proc. 1978 Annu. Conf., 1978, pp. 194-203.
[24] D. B. Shmoys, C. Stein, and J. Wein, "Improved approximation algorithms for shop scheduling problems," inProc. 2nd Annu. SIAM Symp. Discrete Algorithms, 1991, pp. 148-157.
[25] S. L. Tanimoto, "Towards hierarchical cellular logic: Design considerations for pyramid machines," Tech. Rep. 81-02-01, Dep. of Comput. Sci., Univ. of Washington, Feb. 1981.
[26] C. D. Thompson, "A complexity theory for VLSI," Ph.D. dissertation, Dep. Comput. Sci., Carnegie Mellon Univ., 1980.

Index Terms:
multiprocessor interconnection networks; parallel architectures; fat pyramid; universal parallel computation; wire delay; processor size; wire length; simulation; unit wire delay; fat-tree; parallel computation; routing networks; universality.
Citation:
R.I. Greenberg, "The Fat-Pyramid and Universal Parallel Computation Independent of Wire Delay," IEEE Transactions on Computers, vol. 43, no. 12, pp. 1358-1364, Dec. 1994, doi:10.1109/12.338095
Usage of this product signifies your acceptance of the Terms of Use.