Font Size: a A A

Research On SLA Load Balancing And Energy Consumption Oriented Cloud Computing Resource Scheduling

Posted on:2016-10-02Degree:MasterType:Thesis
Country:ChinaCandidate:H R LiuFull Text:PDF
GTID:2308330479989701Subject:Computer Science and Technology
Abstract/Summary:PDF Full Text Request
With the rise of cloud computing and cloud application, especially in recent years, by the popularity of social network, makeing the study of cloud computing is more and more attention by researchers. Because of cloud computing devices are generally very large, the consumption of electricity is also very big, therefore, the energy consumption of the cloud platform and load balancing problems has been widely researched. Cloud computing as a user-oriented platform at the same time, cloud Service providers and users need to sign the SLA to ensure their own interests. According to the definition of service quality, customer by providing feedback for evaluating the quality of service is an important research direction.However, most of the existing research is aimed at a single performance indicators, there are very few to integrate multiple targets within a framework for research. In addition to the user satisfaction and energy consumption, at present most of the research has not considered the effects of the execution time of the task. Peple often think that if you need to ensure the high quality of customer satisfaction, requires the cloud service providers to provide more resources and consuming more energy. Task in view of the partial relaxation model, however, can be extended in guarantee under the premise of customer satisfaction task execution time, in order to save the energy consumption of the entire cloud platform.Therefore, in view of the above problems, this topic propose in view of the multi-objective dynamic resource scheduling strategy, combining the user satisfaction, load balance, comprehensive energy consumption in a frame under consideration, the cloud platform in the case of as much as p ossible to meet the user’s satisfaction, reduce energy consumption, and make the control of load balancing in an acceptable range, improving the resource utilization of the system. To reflect the impact of task execution time for the user satisfaction, thi s subject gives the general user satisfaction and individual tasks user satisfaction, the concept of customer satisfaction. According to the concept of users’ satisfaction in a single task which put forward by this topic, we propose a framework SLSC of res ource scheduling in cloud computing environments. SLSC includes task prediction model and a resource demand forecasting model. In addition, we put forward an algorithm on optimization task model parameters, to provide the more reliable reference for the back of the scheduling algorithm. In addition, in order to balance the relationship between the user satisfaction and energy consumption, this topic proposes an algorithm of the trade-off, in the case of loss minimum user satisfaction, optimize the energy consumption of cloud platform. Finally the experimental results show that prediction algorithmthis topic proposed is accurate and effective, the resource scheduling framework SLSC and trade-off algorithmthis topic proposed can effectively save energy consumption and guarantee customer satisfaction.
Keywords/Search Tags:cloud computing, SLA, load balance, parallel computing
PDF Full Text Request
Related items