Font Size: a A A

Virtual Machine Placement Strategy Based On Particle Swarm Optimization Algorithm

Posted on:2012-10-02Degree:MasterType:Thesis
Country:ChinaCandidate:Y PeiFull Text:PDF
GTID:2248330371965532Subject:Computer technology
Abstract/Summary:PDF Full Text Request
Since the conception of "cloud computing" was put forward in 2006, it quickly became one of the most hot research spot in IT field and is attracting more and more attention of the industry and the academic. As an new term, there are several different definitions of the Cloud Computing.However, it is generally accepted that the Cloud Computing is a new pay-on-demand business model based on virtualization。It is a new Internet computing model with elasticity, dynamicity and resource sharing as its main features. According to the different layers of the service it supplies, cloud computing is divided into three categories:IaaS (Infrastructure as a Service), PaaS (Platform as a service), SaaS (Software as a Service). For IaaS, all the physical servers in the IDC are virtual ized and managed by the centralized management console. In this way, we could benefit from saving budget, uniform deployment, increasing resource utilization rate and reducing power consumingCurrently, the virtualization technology, represented by VMWare, is more and more advanced and has been in use widely. The industry and academic circles have put forward many physical-virtual mapping programs to achieve the aims mentioned above. Both industry and academic try to maximize the resource utilization rate by global mapping method, which is far from perfect. In industry, these programs are represented by DRS of vSphere. However, almost all the current programs are based on the available resource and the requested resource of the new created virtual machine while pay much less attention to the application of the virtual machines offers. For an example, if an application provides service based on many virtual servers, among which there are much date transmission. If the physical distance between them is very far the delay will may be long. So, decomposing the workflow of the application and deploying them into the virtual machines is the problem this paper tries to solve.
Keywords/Search Tags:Cloud Computing, Virtualization, interior delay, PSO, virtual machine placement
PDF Full Text Request
Related items