Phoenix Cloud : Consolidating Heterogeneous Workloads of Large Organizations on Cloud Computing Platforms

For a large organization, different departments often maintain dedicated cluster systems for different workloads, for example parallel batch jobs or Web services. In this paper, we design and implement an innovative cloud computing system software, Phoenix Cloud, to consolidate heterogeneous workloads of the same organization on the cloud computing platform. For Phoenix Cloud, we propose cooperative resource provision and management polices for the affiliated departments of a large organization to share the cluster system. The simulation experiments show: comparing with the previous solution, Phoenix Cloud significantly decreases the configuration scale of the cluster system through consolidating heterogeneous workloads of the same organization, and at the same time increases the number of completed jobs for parallel workload while provisioning enough resources to Web service with varying load.