Load Balancing Based on Optimization Algorithms: An Overview

Abstract

Combinatorial optimization challenges are rooted in real-life problems, continuous optimization problems, discrete optimization problems and other significant problems in telecommunications which include, for example, routing, design of communication networks and load balancing. Load balancing applies to distributed systems and is used for managing web clusters. It allows to forward the load between web servers, using several scheduling algorithms. The main motivation for the study is the fact that combinatorial optimization problems can be solved by applying optimization algorithms. These algorithms include ant colony optimization (ACO), honey bee (HB) and multi-objective optimization (MOO). ACO and HB algorithms are inspired by the foraging behavior of ants and bees which use the process to locate and gather food. However, these two algorithms have been suggested to handle optimization problems with a single-objective. In this context, ACO and HB have to be adjusted to multiobjective optimization problems. This paper provides a summary of the surveyed optimization algorithms and discusses the adaptations of these three algorithms. This is pursued by a detailed analysis and a comparison of three major scheduling techniques mentioned above, as well as three other, new algorithms (resulting from the combination of the aforementioned techniques) used to efficiently handle load balancing issues.

Hybrid Nearest-Neighbor Ant Colony Optimization Algorithm for Enhancing Load Balancing Task Management

“…The MOO algorithm is used in connection with optimization problems that include more than one conflicting objective function to be optimized simultaneously [32]. Multiobjective problems, such as routing in communication networks [33,34], compressor design [35,36], engineering [37,38], and logistics [39], require a number of objective functions for simultaneous optimization [40].…”