Hybrid Algorithm Based on Swarm Intelligence Techniques for Dynamic Tasks Scheduling in Cloud Computing

Abstract

Cloud computing has its characteristics along with so me important issues that should be handled to improve the performance and increase the efficiency of the cloud platform. These issues are related to resources management, fault tolerance, and security. The purpose of this research is to handle the resource management problem, wh ich is to allocate and schedule virtual mach ines of cloud computing in a way that help providers to reduce makespan time of tasks. In this paper, a hybrid algorith m for dynamic tasks scheduling over cloud's virtual machines is introduced. This hybrid algorith m me rges the behaviors of three effective techniques from the swarm intelligence techniques that are used to find a near optimal solution to d ifficu lt co mb inatorial problems. It exp loits the advantages of ant colony behavior, the behavior of particle swarm and honeybee foraging behavior. Experimental results reinforce the strength of the proposed hybrid algorith m. They also prove that the proposed hybrid algorith m is the best and outperformed ant colony optimization, particle swarm optimization , artificial bee colony and other known algorithms.

Efficient Load Balancing in Cloud Environment Using Improved Spider Monkey Optimization

“…In [25], the authors proposed a loadbalancing method based on the cooperative behavior of fireflies. In [26], the authors proposed a hybrid approach to improve makespan and resource utilization. This method combined three algorithms-ant colony, honey bee, and particle swarm.…”

MACO-MOTS: Modified Ant Colony Optimization for Multi Objective Task Scheduling in Cloud Environment

“…This algorithm has the capability to adopt according to dynamic applications, for example, it can modify its character from an artificial ant to real ant. Comparing to FCFS (First Come First Serve) and RR (Round Robin) algorithm, ACO finds better solutions for travelling salesman problems and reduce makespan [9,17]. During runtime overhead, task load gets increase along with lack of rapid adaptability which results in more execution time and decrease in convergence rate will enact as a major pitfall to basic ACO [10].…”