Optimizing Processing Cost of Cloud Using Genetic Algorithm

Authors: Maushmi Mandal, Prof. Mohit Jain

Certificate: View Certificate

Abstract

The cloud computing involve the computational architecture, storage units, networks and the computational units. These units are used for serving the huge amount of request, therefore the resource utilization and their management is essential aspect of cloud computing. In this presented work the resource optimization is the primary goal of the study. In addition of simulation and modeling of the cloud resource preservation is the next goal of the work. In this context recently contributed articles and the research papers are studied. Most of them are utilizing the optimization techniques for improving the performance of the resources. In other words the scheduling of resources can optimize the performance of the cloud resources with respect to the appeared jobs for scheduling. Therefore the genetic algorithm for cloudlet scheduling is presented in this work. Additionally for simulation and modeling of the proposed technique is performed using the cloudSim simulation tool. That is a JAVA based simulation library for supporting the events of cloud and managing cloud resources. After implementation of the genetic algorithm based scheduling technique the proposed model is compared with the exiting space shared cloud scheduling technique. The performance of both the techniques are evaluated in terms of time and space complexity and for providing the difference about the resource utilization the processing cost of the both techniques are also computed. The results demonstrate the proposed technique is efficient and preserve the cost of process execution effectively

Introduction

In recent years the need of computation is increased much significantly. A number of new devices, users and technological gadgets are appeared that frequently accessing the web and internet based services for computing and storage. In order to handle the increasing traffic load on network and computational servers cloud computing is used. The cloud technology offers the different kinds of storage and computational services. These services and their capabilities are not limited they scale their performance and capabilities according to their need. Thus a number of organizations are utilizing the cloud services. On the other hand cloud resources are much expensive and their running cost is large enough. Thus it is also required to preserve these resources by optimization of the scheduling strategy of cloud schedulers. In this presented work the main aim is to study about the cloud scheduling strategies and investigation of their benefits. In next the aim is focused on design and simulation of cloud computing infrastructure for demonstrating the scheduling effect of the cloud. Thus the proposed work involves the implementation of the genetically inspired algorithm for scheduling the cloud resources for optimizing the performance of computational cloud. The genetic algorithm is a soft computing technique that helps to search or optimize constrains. In addition of that the comparative performance studies among the traditionall

Conclusion

compared to the traditional space shared technique. Thus the technique is less resource expensive and efficient in terms of processing of large jobs.