A Cloud Resource Allocation Method Supporting Sudden and Urgent Demands

摘要

In most emergent circumstances, the information resources are not enough or not provided timely, which needs a comprehensive information technique to support rapid and sufficient resource provision. Cloud computing can promote rapid resource allocation, which provides a practical technical support for emergencies. Its key problem is how to guarantee the timeliness and optimization of resource allocation. This paper proposes a cloud resource allocation method supporting sudden and urgent demands, which can allocate various resources timely and optimally for urgent resource demands. This method rebuilds the new priorities of resource allocation in order to allocate resources earlier for urgent virtual machine requests. And a multi-objective optimization mathematical model is established, which sets the minimum performance match distance between virtual machines and physical machines and the minimum number of physical machines as the goals of resource allocation. Then, a multiple-objective optimization algorithm is used to solve this model. We conduct an experiment to exhibit our method. The initial experimental results show that our method has a certain advantage in reducing the number of the used physical machines and improving the resource utilization.