Publication 1995 Issue No. 2 - February Abstract - Efficient Nonblocking Switching Networks for Interprocessor Communications in Multiprocessor Systems
Efficient Nonblocking Switching Networks for Interprocessor Communications in Multiprocessor Systems
February 1995 (vol. 6 no. 2)
pp. 132-141
 ASCII Text x Fong-Chih Shao, A. Yavuz Oruç, "Efficient Nonblocking Switching Networks for Interprocessor Communications in Multiprocessor Systems," IEEE Transactions on Parallel and Distributed Systems, vol. 6, no. 2, pp. 132-141, February, 1995.
 BibTex x @article{ 10.1109/71.342124,author = {Fong-Chih Shao and A. Yavuz Oruç},title = {Efficient Nonblocking Switching Networks for Interprocessor Communications in Multiprocessor Systems},journal ={IEEE Transactions on Parallel and Distributed Systems},volume = {6},number = {2},issn = {1045-9219},year = {1995},pages = {132-141},doi = {http://doi.ieeecomputersociety.org/10.1109/71.342124},publisher = {IEEE Computer Society},address = {Los Alamitos, CA, USA},}
 RefWorks Procite/RefMan/Endnote x TY - JOURJO - IEEE Transactions on Parallel and Distributed SystemsTI - Efficient Nonblocking Switching Networks for Interprocessor Communications in Multiprocessor SystemsIS - 2SN - 1045-9219SP132EP141EPD - 132-141A1 - Fong-Chih Shao, A1 - A. Yavuz Oruç, PY - 1995VL - 6JA - IEEE Transactions on Parallel and Distributed SystemsER -

Abstract—The performance of a multiprocessor system depends heavily on its ability to provide conflict free paths among its processors. In this paper, we explore the possibility of using a nonblocking network with $O\left(N \log N\right)$ edges (crosspoints) to interconnect the processors of an $N$ processor system. We combine Bassalygo and Pinsker's implicit design of strictly nonblocking networks with an explicit construction of expanders to obtain a strictly nonblocking network with $-765.18 N + 352.8 N \log N$ edges and $2+\log \left(N/5\right)$ depth. We present an efficient parallel algorithm for routing connection requests on this network and implement it on three parallel processor topologies. The implementation on a parallel processor whose processing elements are interconnected as in the Bassalygo–Pinsker network requires $O\left(N\log N\right)$ processing elements, $O\left(N\log N\right)$ interprocessor links and it takes $O\left(\log N\right)$ steps to route any single connection request where each step involves a small number ($\approx$72) of bit-level operations. A contracted or folded version of the same implementation reduces the processing element count to $O\left(N\right)$ without increasing the link count or the routing time. Finally, we establish that the same algorithm takes $O\left(\log^3 N\right)$ steps on a perfect shuffle processor with $O\left(N\right)$ processing elements. These results improve the crosspoint, depth and routing time complexities of the previously reported strictly nonblocking networks.

Index Terms—Bassalygo–Pinsker network, Clos network, Cantor network, extensive graph, expander, parallel routing algorithm, strictly nonblocking network.

[1] S. Arora,T. Leighton,, and B. Maggs,“On line algorithms for path selection in a nonblocking network,” Proc. 22nd Ann. ACM Symp. Theory of Computing, pp. 149-158, 1990.
[2] L. Bassalygo,“Asymptotically optimal switching circuits,”Problems of Inform. Transmission, vol. 17, pp. 206–211, July–Sept. 1981.
[3] V.Bene\v s,Mathematical Theory of Connecting Networks and Telephone Traffic. New York: Academic, 1965.
[4] L. Bassalygo and M. Pinsker,“Complexity of optimum nonblocking switching without reconnections,”Problems of Inform. Transmission, vol. 9, pp. 64–66, Jan.–Mar. 1974.
[5] D. G. Cantor,“On non-blocking switching networks,”Networks, vol. 1, pp. 366–377, 1971.
[6] C. Clos,“A study of nonblocking switching networks,”Bell Syst. Tech. J., vol. 32, pp. 406–425, Feb. 1953.
[7] O. Gabber and Z. Galil,“Explicit constructions of linear sized superconcentrators,”J. Comput. Syst. Sci., vol. 22, pp. 407–420, 1981.
[8] C. Y. Lee,“Networks for fast efficient unicast and multicast communications,”Ph.D. Dissertation, Univ. Maryland, Dec. 1992.
[9] G. Lin and N. Pippenger,“Parallel algorithms for routing in nonblocking networks,”inProc. ACM Symp. Theory of Comput., 1991, pp. 272–277.
[10] G. A. Margulis,“Explicit construction of concentrators,”Problems of Inform. Transmission, vol. 9, pp. 325–332, 1973.
[11] D. Nassimi and S. Sahni,“Data broadcasting in SIMD computers,”IEEE Trans. Comput., vol. C-30, pp. 101–107, Feb. 1981.
[12] Z. Galil, N. Alon, and V. D. Milman,“Better expanders and superconcentrators,”J. Algorithms, vol. 8, pp. 337–347, 1987.

Citation:
Fong-Chih Shao, A. Yavuz Oruç, "Efficient Nonblocking Switching Networks for Interprocessor Communications in Multiprocessor Systems," IEEE Transactions on Parallel and Distributed Systems, vol. 6, no. 2, pp. 132-141, Feb. 1995, doi:10.1109/71.342124