Optimizing Static Job Scheduling in a Network of Heterogeneous Computers
ICPP '00 Proceedings of the Proceedings of the 2000 International Conference on Parallel Processing
IEEE Transactions on Computers
Noncooperative load balancing in distributed systems
Journal of Parallel and Distributed Computing
A Nash-Equilibrium Based Algorithm for Scheduling Jobs on a Grid Cluster
WETICE '07 Proceedings of the 16th IEEE International Workshops on Enabling Technologies: Infrastructure for Collaborative Enterprises
Game-Theoretic Approach for Load Balancing in Computational Grids
IEEE Transactions on Parallel and Distributed Systems
Cooperative load balancing in distributed systems
Concurrency and Computation: Practice & Experience
Hi-index | 0.00 |
Load balancing is very important and a common problem in distributed systems. The load balancing mechanism aims to fairly distribute load across the resources so as to optimize a given objective function. The objective can be system optimality which tries to minimize mean response time of all users or individual optimality which tries to minimize each user's individual response time. The load balancing can be achieved either statically or dynamically. In this paper we review, competitive equilibrium (CE) approach for static load balancing and then propose recursive competitive equilibrium (RCE) approach for dynamic load balancing. A computer model is run to evaluate the performance of proposed RCE scheme with static scheme using Nash equilibrium (NE) approach and the static scheme using CE. The results show that static scheme using CE and dynamic scheme using RCE achieved both system optimality and individual optimality simultaneously, while NE scheme achieved only individual optimality. Moreover the performance of RCE scheme is higher than CE when communication overhead is less and almost same as CE scheme when the communication overhead is more.