期刊名称：Journal of Theoretical and Applied Information Technology
印刷版ISSN：1992-8645
电子版ISSN：1817-3195
出版年度：2014
卷号：67
期号：2
出版社：Journal of Theoretical and Applied
摘要：Cloud computing is a new computing paradigm, which offers resources for solving complex problem in fast and lower cost under pay per usage through internet. This feature is great boon to software companies to reduce their infrastructure setup and maintenance cost. IaaS service is one of the fundamental service models of cloud provider that offers entire computing environment as Virtual Machine (VM) to customers. The VM is created on cloud provider datacenter physical server. The provider deployed multiple datacenters in geographically different location to cater the needs of IaaS customers. The customers are much concerned about reducing their computation time and VM rental cost. The cloud service brokers are playing vital role in this regard, one of their responsibility is to direct the user request to appropriate datacenter. The selection of datacenter is a challenging task for service broker. Our proposed approach aims to enrich the intelligence of service broker during datacenter selection. Our novel algorithm makes the service broker to aware about the future load of every datacenter during request forwarding. It would facilitate the broker to route the IaaS request in right destination. The performance of our novel methodology is tested in Cloud Analyst tool. The result shows that our methodology reduces task time of customer applications compare to existing broker policies.
关键词：Virtual Machine (VM); Infrastructure as a Service (IaaS); Datacenters Classification Algorithm (DCA); Datacenter Load Aware Service broker Algorithm (DLASA); Neural Networks (NN).
