
This Article  
 
Share  
Bibliographic References  
Add to:  
Digg Furl Spurl Blink Simpy Del.icio.us Y!MyWeb  
Search  
 
ASCII Text  x  
K.W. Tang, S.A. Padubidri, "Diagonal and Toroidal Mesh Networks," IEEE Transactions on Computers, vol. 43, no. 7, pp. 815826, July, 1994.  
BibTex  x  
@article{ 10.1109/12.293260, author = {K.W. Tang and S.A. Padubidri}, title = {Diagonal and Toroidal Mesh Networks}, journal ={IEEE Transactions on Computers}, volume = {43}, number = {7}, issn = {00189340}, year = {1994}, pages = {815826}, doi = {http://doi.ieeecomputersociety.org/10.1109/12.293260}, publisher = {IEEE Computer Society}, address = {Los Alamitos, CA, USA}, }  
RefWorks Procite/RefMan/Endnote  x  
TY  JOUR JO  IEEE Transactions on Computers TI  Diagonal and Toroidal Mesh Networks IS  7 SN  00189340 SP815 EP826 EPD  815826 A1  K.W. Tang, A1  S.A. Padubidri, PY  1994 KW  hypercube networks; performance evaluation; network routing; toroidal mesh networks; diagonal mesh networks; degree4 point to point interconnection models; parallel computers; multicomputers; wirability; scalability; optimal selfrouting algorithm; performance; deflection routing. VL  43 JA  IEEE Transactions on Computers ER   
Diagonal and toroidal mesh are degree4 point to point interconnection models suitable for connecting communication elements in parallel computers, particularly multicomputers. The two networks have a similar structure. The toroidal mesh is popular and wellstudied whereas the diagonal mesh is relatively new. In this paper, we show that the diagonal mesh has a smaller diameter and a larger bisection width. It also retains advantages such as a simple rectangular structure, wirability and scalability of the toroidal mesh network. An optimal selfrouting algorithm is developed for these networks. Using this algorithm and the existing routing algorithm for the toroidal mesh, we simulated and compare the performance of these two networks with N=35/spl times/71=2485, N=49/spl times/99=4851, and N=69/spl times/139=9591 nodes under a constant system with a fixed number of messages. Deflection routing is used to resolve conflicts. The effects of various deflection criteria are also investigated. We show that the diagonal mesh outperforms the toroidal mesh in all cases, and thus provides an attractive alternative to the toroidal mesh network.
[1] T. Y. Feng, "A survey of interconnection networks,"Computer, vol. 14, pp. 1227, Dec. 1981.
[2] L. D. Wittie, "Communication structures for large networks of microcomputers,"IEEE Trans. Comput., vol. 30, no. 4, pp. 264273, Apr. 1981.
[3] F. T. Leighton,Introduction to Parallel Algorithms and Architectures: Arrays, Trees, and Hypercubes. Palo Alto, CA: Morgan Kaufmann, 1992.
[4] C. D. Thompson, "Areatime complexity for VLSI," inProc. Eleventh Annu. ACM Symp. Theory Comput., 1979, pp. 8188.
[5] M. BenAyed, "Dynamic routing for regular direct computer networks," Ph.D. thesis, Elec. Eng. Dep., College of Eng. and Appl. Sci., Univ. of Rochester, Rochester, NY, 1990.
[6] N. F. Maxemchuk, "Routing in the Manhattan Street Network,"IEEE Trans. Commun., vol. Com35, no. 5, May 1987, pp. 503512.
[7] N. F. Maxemchuk, "Comparison of deflection and storeandforward techniques in the Manhattan street and shuffleexchange networks," inProc. IEEE INFOCOM'89, 1989, pp. 800809.
[8] J. C. Bermond, C. Delorme, and J. J. Quisquater, "Strategies for interconnection networks: Some methods from graph theory,"J. Parallel and Distributed Comput., vol. 3, pp. 433449, 1986.
[9] K.W. Tang, "Dense symmetric interconnection networks," Ph.D. thesis, Elec. Eng. Dep., College of Eng. and Appl. Sci., Univ. of Rochester, Rochester, NY, 1991.
[10] B. W. Arden, private communication, 1991.
[11] B. W. Arden and F. Li, "Simulation of interconnections for massively parallel systems," Tech. Rep., Dep. of Elec. Eng., Univ. of Rochester, Rochester, NY, 1991.
[12] K. W. Tang and S. A. Padubidri, "Routing and diameter analysis of diagonal mesh networks," InProc. Int. Conf. Parallel Processing, St. Charles, IL, Aug. 1721 1992, pp. I143150.
[13] Almasi, G.S., and Gottlieb, A.,Highly Parallel Computing, Benjamin/Cummings, Redwood City, Calif., 1989 [survey text: parallel architectures (Connection Machine, GF11, hypercubes, sharedmemory multiprocessors, Ultra, RP3, Cedar), programming difficulties, and compiler methods].
[14] D. Parkinson and J. Litt, Eds.,Massively Parallel Computing with the DAP. Cambridge, MA: M.I.T Press, 1990.
[15] N. F. Maxemchuk, "Regular mesh topologies in local and metropolitan area networks,"AT&T Tech. J., vol. 64, no. 7, pp. 16591685, Sept. 1985.
[16] F. Borgonovo and E. Cadarin, "Routing in the bidirectional Manhattan network," inProc. Third. Conf. Data Commun. Syst. and their Perform., Rio de Janerio, June 1987, pp. 181189.
[17] K. W. Tang and B. W. Arden, "Vertextransitivity and routing for Cayley graphs in GCR representations," inProc. 1992 Symp. Appl. Computing, Kansas City, MO, Mar. 13, 1992, pp. 11801187.
[18] Baran, P. 1964. On distributed communications networks.IEEE Trans. on Communications Systems COM12, pp. 19.
[19] W. D. Hillis,The Connection Machine. Cambridge, MA: MIT Press, 1985.
[20] A. Krishna and B. Hajek, "Performance of shufflelike switching networks with deflection," inProc. IEEE INFOCOM'90, San Francisco, CA, June 1990, pp. 473480.
[21] M. Decina, P. Giacoazzi, and A. Pattavina, "Shuffle interconnection networks with deflection routing for ATM switching: The closedloop shuffleout," inProc. IEEE INFOCOM'91, Bal Harbour, FL, Apr. 1991, pp. 12531263.