Research On Virtual Machine Dynamic Management Based On OpenStack Platform

In the modern society, along with information quantity and data processing quantity unceasing increase, a new model of computing--cloud computing came into being. Cloud computing is a new service mode which combines the parallel computing, grid computing, utility computing and distributed storage technology. With the explosive development of cloud computing, resource management problems also emerge in an endless stream, has become the key to influence the development of cloud computing. Virtual machine management and scheduling is the most important resources management in cloud platform. From the perspective that effective utilization of various computing resources in park cloud environment. Proposed the dynamic task scheduling algorithm based on AHP, improve the park cloud platform for various physical resource utilization effectively.In this scenario, we proposed a dynamic management scheme based on AHP and concerns of the unified management and qualified about running status of virtual machine.The main work of this paper includes the following aspects:1. The Cloud Computing Technology was introduced with the research, and a detailed analysis of the present popular cloud computing architecture technology. Finally select combine Collected with Graphite that suitable for monitoring on Openstack cloud platform and successful build a cloud environments programme;2. In the cloud environment, combining with the existing storage technology, using the iSCSI protocol, first proposed multi-node block storage project and complete the deployment and experiment.3. Combined with own API interface, using the shared storage technology NFS, complete dynamic live-migration of virtual machine between the virtual across nodes.4. Combined with the requirements of tenants and maximum using of hardware resources on platform, designed a virtual machine dynamic management scheme bas- ed on AHP analysis and simulation tests have been conducted. This paper proposed a dynamic management scheme of virtual machine based on AHP analysis has certain application value. Meet the requirements for dynamic management of virtual machine resources and real-time demand of the tenant. At the same time, it has certain reference value for large-scale public cloud computing environment to improve the user experience.