Research on Resource Management in PaaS Based on IaaS Environment - Frontiers in Internet Technologies - page 156

6 Conclusion and Future Works

It's becoming a trend for enterprises to adopt PoI. In summary, PoI has many advan-
tages on the flexibility, reliability and cost-efficiency. In this paper, a QoS guaranteed
and cost-efficient resource management framework in PoI is proposed. The frame-
work is composed of a feedback control system and two scaling algorithms. Experi-
mental results show that the resource management framework can not only maintain
Missed Deadline Ratio of all applications at an expected value, but also improve the
CPU Utilization of all applications around an expected value. According to experi-
mental results, it can be found that Time Usage of VMs has negative correlation with
the Set Point of Missed Deadline Ratio, the Set Point of CPU Utilization and VM
Safe Threshold.
In this paper, the resource management framework is designed based on a relative-
ly simplified IaaS Layer, e.g., the action of renting and returning VMs can be finished
immediately. However this assumption usually does not hold true. For example,
Amazon Elastic Compute Cloud (Amazon EC2), an IaaS provider, charges on hourly
basis, which makes it impossible to return VMs to IaaS providers immediately. In the
future, a more practical abstraction of IaaS Layer will be involved, e.g., the potential
delay in renting and returning VMs needs to be taken into account.

References

1. Chandra, A., Gong, W., Shenoy, P.: Dynamic Resource Allocation for Shared data centers
using online measurements. In: Proceedings of the 11th International Workshop on Quality
of Service (2003)
2. Ruth, P., McGachey, P., Xu, D.: VioCluster, “Virtualization for Dynamic Computational
Domains”. IEEE International on Cluster Computing, 1-10 (September 2005)
3. Menasc, D., Casalicchio, E.: A Framework for Resource Allocation in Grid Computing.
In: Proceedings of the 12th Annual International Symposium on Modeling, Analysis, and
Simulation of Computer and Telecommunications Systems, pp. 259-267 (2004)
4. Yazir, Y., Matthews, C., Farahbod, R., Neville, S., et al.: Dynamic Resource Allocation in
Computing Clouds using Distributed Multiple Criteria Decision Analysis. In: 3rd Interna-
tional Conference on Cloud Computing, Miami, Florida, USA (2010)
5. Chang, F., Ren, J., Viswanathan, R.: Optimal Resource Allocation in Clouds. In: 3rd Inter-
national Conference on Cloud Computing, Miami, Florida, USA (2010)
6. Mazzucco, M., Dyachuk, D., Deters, R.: Maximizing Cloud Providers Revenues via Ener-
gy Aware Allocation Policies. In: 3rd International Conference on Cloud Computing, Mi-
ami, Florida, USA (2010)
7. Bossche, R., Vanmechelen, K., Broeckhove, J.: Cost-Optimal Scheduling in Hybrid IaaS
Clouds for Deadline Constrained Workloads. In: 3rd International Conference on Cloud
Computing, Miami, Florida, USA (2010)
8. Cristian, F., Fetzer, C.: The Timed Asynchronous Distributed System Model. IEEE Trans-
actions on Parallel and Distributed Systems (June 1999)
9. Hu, R., Li, Y., Zhang, Y.: Adaptive Resource Management in PaaS Platform Using Feed-
back Control LRU Algorithm. In: 2011 International Conference on Cloud and Service
Computing (2011)