The effect of time delays on the stability of load balancing algorithms for parallel computations

A deterministic dynamic nonlinear time-delay system is developed to model load balancing in a cluster of computer nodes used for parallel computations. The model is shown to be self consistent in that the queue lengths cannot go negative and the total number of tasks in all the queues and the network are conserved (i.e., load balancing can neither create nor lose tasks). Further, it is shown that using the proposed load balancing algorithms, the system is stable in the sense of Lyapunov. Experimental results are presented and compared with the predicted results from the analytical model. In particular, simulations of the models are compared with an experimental implementation of the load balancing algorithm on a distributed computing network.

[1]  Anthony P. Reeves,et al.  Strategies for Dynamic Load Balancing on Highly Parallel Computers , 1993, IEEE Trans. Parallel Distributed Syst..

[2]  J. Douglas Birdwell,et al.  A HIERARCHICAL DATABASE DESIGN AND SEARCH METHOD FOR CODIS , 2000 .

[3]  E. Jury,et al.  Stability independent and dependent of delay for delay differential systems , 1984 .

[4]  Tse-Wei Wang,et al.  Dynamic Time Delay Models for Load Balancing. Part I: Deterministic Models , 2004 .

[5]  Jie Li,et al.  A performance comparison of dynamic vs. static load balancing policies in a mainframe-personal computer network model , 2000, Proceedings of the 39th IEEE Conference on Decision and Control (Cat. No.00CH37187).

[6]  L. Schrage,et al.  Queueing systems, Vol. I: Theory , 1977, Proceedings of the IEEE.

[7]  François Spies Modeling of optimal load balancing strategy using queueing theory , 1996, Microprocess. Microprogramming.

[8]  E. W. Kamen,et al.  Linear systems with commensurate time delays: stability and stabilization independent of delay , 1982 .

[9]  Tse-Wei Wang,et al.  The effect of time delays in the stability of load balancing algorithms for parallel computations , 2003, 42nd IEEE International Conference on Decision and Control (IEEE Cat. No.03CH37475).

[10]  Sagar Dhakal,et al.  Dynamical discrete-time load balancing in distributed systems in the presence of time delays , 2003, 42nd IEEE International Conference on Decision and Control (IEEE Cat. No.03CH37475).

[11]  Majeed M. Hayat,et al.  On Load balancing in distributed systems with large time delays: Theory and experiment , 2004 .

[12]  Majeed M. Hayat,et al.  The effect of feedback gains on the perfomance of a load balancing network with time delays , 2003 .

[13]  Franco Zambonelli,et al.  Diffusive load-balancing policies for dynamic applications , 1999, IEEE Concurr..

[14]  亀田 壽夫,et al.  Optimal load balancing in distributed computer systems , 1997 .

[15]  H. G. Rotithor Taxonomy of dynamic task scheduling schemes in distributed computing systems , 1994 .

[16]  Tse-Wei Wang,et al.  Load balancing instabilities due to time delays in parallel computations , 2001 .

[17]  J. Douglas Birdwell,et al.  Timing and Congestion Studies under PVM , 2001, PPSC.

[18]  Eitan Altman,et al.  Equilibria for multiclass routing in multi-agent networks , 2001, Proceedings of the 40th IEEE Conference on Decision and Control (Cat. No.01CH37228).

[19]  John Chiasson,et al.  A method for computing the interval of delay values for which a differential-delay system , 1986 .