This Article 
   
 Share 
   
 Bibliographic References 
   
 Add to: 
 
Digg
Furl
Spurl
Blink
Simpy
Google
Del.icio.us
Y!MyWeb
 
 Search 
   
Performance Modeling of Hierarchical Crossbar-Based Multicomputer Systems
September 2001 (vol. 50 no. 9)
pp. 877-890

Abstract—Crossbar networks have been widely used as the interconnection mechanism in many multiprocessor/multicomputer systems. Crossbar networks can be categorized into three major topological classes: full-crossbar networks, multistage interconnection networks (MINs), and networks consisting of multiple levels of full crossbar connections, called hierarchical crossbar interconnection networks (HCINs). A significant amount of previous work exists in the area of performance modeling of systems with full-crossbar networks or multistage interconnection networks. However, performance modeling of multicomputer systems with HCINs has not been widely studied. This paper presents both analytical and simulation models for performance evaluation of an HCIN based on the commercial Mercury RACEway crossbar switch. The effective data transfer rate for message passing is taken as the primary performance metric and the models predict how this metric varies with the traffic load on the system. The analytical results are compared to the simulation results for different standard configurations of Mercury RACE Multicomputer Systems.

[1] T. Feng, “A Survey of Interconnection Networks,” Computer, vol. 14, no. 12, pp. 12-27, Dec. 1981.
[2] J.H. Patel, “Performance of Processor-Memory Interconnections for Multiprocessors,” IEEE Trans. Computers, vol. 30, no. 10, pp. 771-780, Oct. 1981.
[3] I-Cube Digital Crosspoint Switches (DCS),http:/www.icube.com, 2001.
[4] Applied Micro Circuits Corporation, Crosspoint Switches,http:/www.amcc.com, 2001.
[5] Vitesse Semiconductor Corporation, VSC880 Crosspoint Switch,http:/www.vitesse.com, 2001.
[6] Pericom Semiconductor Corporation, SiliconSwitch,http:/www.pericom.com, 2001.
[7] N. Pippenger, “On Crossbar Switching Networks,” IEEE Trans. Comm., vol. 23, no. 6, pp. 646-659, June 1975.
[8] L.N. Bhuyan and D.P. Agrawal, “Design and Performance of Generalized Interconnection Networks,” IEEE Trans. Computers, vol. 32, no. 12, pp. 1081-1090, Dec. 1983.
[9] L.N. Bhuyan,Q. Yang, and D.P. Agrawal,"Performance of Multiprocessor Interconnection Networks," Computer, vol. 22, no. 2, pp. 25-37, Feb. 1989.
[10] E.W. Zegura, “Evaluating Blocking Probability in Generalized Connectors,” IEEE/ACM Trans. Networking, vol. 3, no. 4, pp. 387-398, Aug. 1995.
[11] A.K. Ramani, P.K. Chande, and P.C. Sharma, “A General Model for Performance Investigations of Priority Based Multiprocessor System,” IEEE Trans. Computers, vol. 41, no. 6, pp. 747-754, June 1992.
[12] F.T. Hady and B.L. Menezes, “The Performance of Crossbar-Based Binary Hypercubes,” IEEE Trans. Computers, vol. 44, no. 10, pp. 1208-1215, Oct. 1995.
[13] S.M. Mahmud,"Performance of Multilevel Bus Networks for Hierarchical Multiprocessors," IEEE Trans. Computers, vol. 43, no. 7, pp. 789-805, July 1994.
[14] The RACE Multicomputer, Volume 1, Version 1.3, Mercury Computer Systems Inc., 1995.
[15] C.E. Leiserson, "Fat-Trees: Universal Networks for Hardware Efficient Supercomputing," IEEE Trans. Computers, vol. C-34, no. 10, Oct. 1985, pp. 892-901.
[16] R.I. Greenberg and C.E. Leiserson, “Randomized Routing on Fat-Trees,” Proc. IEEE 26th Ann. Symp. Foundations of Computer Science, Nov. 1985.
[17] A. DeHon, “Practical Schemes for Fat-Tree Network Construction,” MIT Transit Project, Transit Note #32, 1993.
[18] C.T. Lea,“Multi-log2N networks and their applications inhigh speed electronic and photonic switching systems,” IEEE Trans. Communications, vol. 38, pp. 1,740-1,749, 1990.
[19] M. Holliday and M. Stumm, “Performance Evaluation of Hierarchical Ring-Based Shared Memory Multiprocessors,” IEEE Trans. Computers, vol. 43, no. 1, pp. 52-67, Jan. 1994.
[20] S. Kumar, R.H. Klenke, J.H. Aylor, B.W. Johnson, R.D. Williams, and R. Waxman, ADEPT: A Unified Environment for End-to-End System Design, in High-Level System Modeling: Specification and Design Methodologies, pp. 55-82. Kluwer Academic, 1996.
[21] K. Prasad and A. Singhal, “Simulation of Ethernet Performance Based on Single Server and Single Queue Model,” Proc. 12th Conf. Local Computer Networks, pp. 74-85, 1987.
[22] C.G. Cassandras, Discrete Event Systems, Modeling and Performance Analysis. Homewood, Ill.: Richard D. Irwin, Inc., and Aksen Assoc., Inc., 1993.
[23] L. Kleinrock, Communication Nets: Stochastic Message Flow and Delay. New York: McGraw-Hill, 1964.
[24] R.H. Klenke, J.H. Aylor, R. Hillson, and D.J. Kaplan, “VHDL-Based Performance Modeling for the Processing Graph Method Tool (PGMT) Environment,” Proc. VHDL Int'l Users Forum, pp. 69-78, Feb. 1996.
[25] G. Han, R.H. Klenke, and J.H. Aylor, “Performance Modeling of Hierarchical Crossbar-Based Multicomputer Networks,” Proc. IASTED Conf. Modeling and Simulation, May 1999.
[26] R. Hillson, “Support Tools for the Processing Graph Method,” Proc. Fifth Int'l Conf. Signal Processing Applications and Technology (ICSPAT 94), pp. 756-761, Oct. 1994.
[27] Mentor Graphics, QuickHDL User's and Reference Manual, Software Version 8.5_4, Appendix A, VHDL Foreign Language Interface, 1995.
[28] MATLAB(R), Version 5.1.0.421. Natick, Mass.: The Mathworks, Inc., 1999.

Index Terms:
Multicomputer systems, crossbar interconnection networks, performance evaluation, modeling, simulation.
Citation:
Gang Han, Robert H. Klenke, James H. Aylor, "Performance Modeling of Hierarchical Crossbar-Based Multicomputer Systems," IEEE Transactions on Computers, vol. 50, no. 9, pp. 877-890, Sept. 2001, doi:10.1109/12.954504
Usage of this product signifies your acceptance of the Terms of Use.