Energy-Efficient Independent Task Scheduling In Cloud Computing

Cloud Computing is a technology that permits users to host their applications or enroll in computing resources on far off servers in a pay-as-you-goal model.The customers get access to those applications and computing sources within the shape of internet services.These offerings supplied by means of the Cloud providers may be extensively categorized as: software-as-a-service(SaaS),Platform-as-a-carrier(PaaS)and Infrastructureas-a-provider(IaaS).But energy consumption is one of the most important challenge in cloud computing as large amount of energy is wasted through the data centres web hosting these cloud applications.The high scientific applications,contain thousands of tasks are being to be execute in virtulized cloud for lots benefits.Increase the processing capability of the cloud system for execute these tasks,the computation power is also increased substantially.How to reduce the energy consumption from cloud computing data center has become an urgent problem in academia and industry.In this thesis,we focus on the issues of scheduling independent tasks in a heterogeneous computing environment,considering minimization of the overall computation energy.We did prefer to propose a green computing solution which isn't always only able to minimize computation energy however further to reduce the environmental pollutants.In order to solve these issues,it is easy to solve if we schedule the task to appropriate VM.We schedule the tasks to the slowest virtual machine,fulfill the deadline while minimizing the energy consumption.Based on involved characteristics and properties of the considered problem,we build the mathematical model for the optimization goal and constraints then a heuristic algorithm is proposed,the heuristic algorithm consists of three component:task sequencing,VM searching and Task mapping.In order to verify the performance of the proposed algorithm,this paper designs two experimental modules for parameter calibration and algorithm comparison.Cloudsim a toolkit for modeling and simulation of cloud computing,has been used to put implement and show the experimental effects.Experimental results indicate that the proposed algorithm has better performance than other benchmark algorithms under different deadline candidates.