Workload Modeling and Scalability Analysis of Resource Usage for E-commerce Application

Abstract

Performance analysis and simulation of cloud resource management policies allows cloud providers to improve their QOS. QOS denotes the level of performance, reliability and availability offered by the web application and by the platform infrastructure that hosted it. The modeling of workload is difficult due to the unavailability of trace logs for analysis. A web application model has been suggested to capture different behavior of user profile. The amount of the resource consumed during the process is known as resource usage. A resource usage model shall be created and validated to choose the best estimated policy among Generalized Lambda distribution & Generalized Extreme Value distribution. The validation is carried out by simulating the same model in cloudsim. The obtained results from amazon web services environment and cloudsim simulator are compared by Wilcox method and the accuracy is defined.