Thesis (M.Sc.) - Cairo University - Faculty of Computers and Information - Department of Computer Science

In the last few years, the Cloud Computing has become the fast spread in the field of computing. According to the Cloud Computing, there are new possibilities for building applications and providing various services to the end user by virtualization through the internet. On the other hand, the task scheduling problem is one of the most significant challenges in the Cloud Computing because the user has to pay for the needed resources on the basis of time, which acts to distribute the load evenly among the system resources by maximizing utilization and, besides reducing task execution time.The Particle Swarm Optimization (PSO), Genetic Algorithm (GA), Ant Colony Optimization (ACO) and Tabu search (TS) are important heuristic algorithms and/or approaches for solving several problems. Task Scheduling is one of such problems In spite of the Particle Swarm Optimization (PSO) algorithm is considered, a simple parallel algorithm can be applied in different ways to resolve the task scheduling problems. This involves two main drawbacks mainly :1. The initial population is randomly selected which leads to go far the best solution. 2. The weakness of the local searches because there is a possibility to be trapped in a local search in the last repetition process. According to the work in this thesis, two modified task scheduling algorithms have been introduced based on PSO to overcome its drawbacks. According to the first proposed algorithm, the Best- Fit (BF) algorithm has been merged into the standard PSO algorithm to generate the initial population of the standard PSO algorithm to obtain a good initial selection. This algorithm is called BFPSO

Issued also as CD