A Comparative Study of the Performance of Load Balancing Algorithms Using Cloud Analyst

Abstract

Cloud computing technology has grown and expanded widely in recent years. Cloud computing is a distributed system that may be on physical or virtual machines. Because of the ease of working in the cloud, its ease of use, it's being an interactive model and a highly profitable business model, many users have used it. This led to an increase in the load on it, and thus the overload has become a problem that must be solved. Many algorithms have been proposed to balance the load on cloud. The essential are three: Round robin, Equally Spread Current Execution (ESCE), and Throttled. This research investigates the performance of those algorithms with increasing of number of users. Previous researches did not take into account the increasing of users. CloudAnalyst simulation tool is used for this comparison. The comparison is done according to DC (Data Centre) processing time, overall total response time (ORT), processing cost, and hourly data center. It was found that initially with a certain settings and for a certain number of users, there may be no difference that can be mentioned in performance between the three algorithms. Then the improvement in performance increases with the increasing number of users. Throttled algorithm showed best performance in the average of ORT with increasing number of users.