This Article 
   
 Share 
   
 Bibliographic References 
   
 Add to: 
 
Digg
Furl
Spurl
Blink
Simpy
Google
Del.icio.us
Y!MyWeb
 
 Search 
   
The Probability of Load Balancing Success in a Homogeneous Network
September 1991 (vol. 17 no. 9)
pp. 922-933

The problem of load balancing in distributed systems composed of several homogeneous sites connected by a subnet is examined. The author determines a general formula for the probability that any one site in the system is underloaded while some other site in the system is overloaded. This probability can be used to define the likelihood of load balancing success in a distributed operating system. This probability gives insight into the utilization of the system and is an aid in determining a measure of effectiveness of the system. From this formula one can determine this probability when the workload is composed of processes typical to distributed systems. The influence of variants in the load balancing algorithm on this probability is demonstrated.

[1] R. Byrant and R. Finkel, "A stable distributed scheduling algorithm," inProc. IEEE Int. Conf. Distributed Comput. Syst., 1981, pp. 314-324.
[2] G. Burns and R. Daoud, "The performance / functionality dilemma of multicomputer message passing," inProc. IEEE DMCC5 Conf.(Charleston, SC), 1990, to be published.
[3] D. R. Cheriton, "The V kernel distributed operating system,"Commun. ACM, vol. 1, pp. 105-115, Feb. 1985.
[4] D. R. Cheriton and W. Zwaenepol, "The distributed v kernel and its performance for diskless workstations," inProc. 9th Symp. Operating System Principles, ACM SIGOPS, Bretton Woods, NH, Oct. 1983, pp. 128-139.
[5] D. Eager, E. Lazowska, and J. Zahorjan, "Adaptive load sharing in homogeneous distributed systems,"IEEE Trans. Software Eng., vol. SE-12, no. 5, pp. 662-675, May 1986.
[6] D. L. Eager, E. Lazowska, and J. Zahorjan, "A comparison of receiver-initiated and sender-initiated adaptive loading,"Perform. Eval., vol. 6, 1986.
[7] K. Efe, "Heuristic models of task assignment scheduling in distributed systems,"IEEE Computers, vol. C-15, pp. 50-56, June 1982.
[8] A. Ezzat, "Load balancing in nest: a network of workstations," inProc. IEEE Fall Joint Computer Conf., 1986, pp. 1138-1148.
[9] D. Farberet al., "The distributed computer system," inProc. 7th Ann. IEEE Computer Soc. Int., 1973, pp. 31-34.
[10] D. Grunwald, B. Nazief, and D. Reed, "Empirical comparison of heuristic load distributions in point-to-point multicomputer networks," inIEEE DMCC5 Conf.(Charleston, SC), Apr. 8-12, 1990, pp. 984-993.
[11] M. Gross and C. Harris,Introduction to Queueing Theory. New York: Wiley, 1976.
[12] D. Hinz, "A run-time load balancing strategy for highly parallel systems,"IEEE DMCC5 Conf.(Charleston, SC), Apr. 8-12, 1990, pp. 951-961.
[13] L. Kleinrock,Queueing Systems, vol. 1. New York: Wiley, 1976.
[14] B. A. Kingsbury, "The network queueing system," NASA Ames, Moffett Field, CA, Rep. 11786, 1986.
[15] W. E. Leland and T. J. Ott, "Load-balancing heuristics and process behavior," inProc. 1986 ACM Sigmetrics Conf. Measurement Modeling Comput. Syst. Perform. Eval. Rev., vol. 14, no. 1, pp. 54-69, North Carolina State Univ., May 1986.
[16] M. Livny and M. Melmen, "Load balancing in homogeneous broadcast distributed systems," inProc. Computer Network Perform. Symp., 1982, pp. 47-55.
[17] D. May, "OCCAM,"SIGPLAN Notices, vol. 18, no. 4, pp. 69-79, Apr. 1983.
[18] R. Mirchandaney, D. Towsley, and J. Stankovic, "Analysis of the effects of delays on load sharing," Univ. Massachusetts, Amherst, COINS Tech. Rep. 87-100, 1987.
[19] K. Ramamritham and J. Stankovic, "Dynamic task scheduling in hard real-time distributed systems,"IEEE Computers, vol. C-1, pp. 96-107, Jan. 1984.
[20] C. G. Rommel, "Scheduling parallel programs in distributed systems," Ph.D. thesis, Univ. Massachusetts, Amherst, Sept. 1988.
[21] C. G. Rommel, "The probability of load balancing success," inProc. ISMM Conf.(Fort Lauderdale, FL), Dec. 13-15, 1989, pp. 18-22.
[22] C. G. Rommel and R. Rossow,XNOS Users Reference Manual. Hartford, CT: United Technol. Res. Ctr., 1989.
[23] T. Saaty,Elements to Queueing Theory. New York: McGraw-Hill, 1961.
[24] C. L. Seitz, "The Cosmic Cube,"Commun. ACM, pp. 22-33, Jan. 1985.
[25] J. Stankovic, "An application of Bayesian decision theory to decentral control of scheduling,"IEEE Trans. Computers, vol. C-3, pp. 117-130, Feb. 1985.
[26] D. Sweetman and J Munoz, "Measures of effectiveness and the performance of parallel architectures," presented at the Systems, Man, Cybern. Conf., 1989.
[27] Parallel Systems user's Guide., Topologix, Inc., Denver, CO, 1989.

Index Terms:
homogeneous sites; subnet; general formula; distributed operating system; load balancing algorithm; computer networks; distributed processing; operating systems (computers); probability
Citation:
C.G. Rommen, "The Probability of Load Balancing Success in a Homogeneous Network," IEEE Transactions on Software Engineering, vol. 17, no. 9, pp. 922-933, Sept. 1991, doi:10.1109/32.92912
Usage of this product signifies your acceptance of the Terms of Use.