Application software service is foundation in science cloud computing, but it's always related to limited software licenses and hardware resources. So, it is important to study the software and hardware co-allocation algorithm and strategy under the environment of science cloud computing. In this paper, a novel architecture of campus science cloud computing is proposed which can support policy-based software and hardware management and allocation. Then, a model is set up considering resources, users, jobs and applications. Furthermore, the software and hardware co-allocation algorithm SHA is proposed. The algorithm SHA has been used in the NPU campus cloud computing environment. The results show that it can improve the utilization rate of software, realize the difference of QoS between the users with different roles, and ensure the fairness between the users with the same role.
Software And Hardware