Design And Implementation Of Dynamic Resource Allocation Framework Based On OpenStack

Cloud computing has become the most important and popular topics in the IT industry undoubtedly. It not only brings a revolutionary innovation for the use and management of IT resources. With the increase of cloud users, how to make virtual resources in the cloud data center remove redundant data on the basis of the realization of disaster, how to reduce users’ waiting time and improve the resource utilization at the same time to make the resources in the cloud data center used fully, have become one of the most important problems of resource schedule about cloud computing environment.This paper mainly studies the dynamic scheduling strategy of the virtual computing resources in the IaaS cloud platform. It bases on OpenStack which is the most famous open source IaaS cloud platform currently. The main studies and innovations are as follows:(1) Analyze the development background and the significance of cloud computing technology at home and abroad and make a detailed presentation of the definition of cloud computing, the basic characteristics and the types of service. Study the OpenStack resource scheduling mechanism which contains the resource scheduling module and the resource scheduling process.(2) Quantify the virtual computing resources for IaaS cloud platform based on OpenStack. Propose and design the computing resource scheduling algorithms on the OpenStack cloud platform, including CPU-intensive, memory-intensive scheduling algorithms and real-time feedback integrated load balancing resource optimization strategy and so on. Meanwhile, elaborate the optimize real-time feedback API python concept and the definition of the XML file.(3) Propose a scalable distributed resource scheduling multiple resource pools framework based on OpenStack by analyzing the current resource scheduling mechanism of the mainstream open source IaaS cloud platform. This framework divides the physical host into multiple resource pools, set the global optimization strategy for the global resources and set the local optimization strategy in each resource pool itself. Experimental results show that the distributed muti-resource pool proposed in this paper achieves a good resource optimization, improves the resource utilization, provides the performance of the isolation and security isolation and has a good scalability at the same time.