Cost Optimization for Time-Bounded Request Scheduling in Geo-Distributed Datacenters
To cope with the growing service requests, a large number of cloud services are deployed in geographically distributed datacenters for better performance. Currently, how to optimize the monetary expenditure spent on VM (Virtual Machine) rental has been widely concerned. Especially, the diversities of the rental prices and service capabilities in geo-distributed regions make the problem more complicated. In this paper, the time restriction of requests and load balance are both taken into account when optimizing the VM rental cost. A two-layer geo-distributed request scheduling algorithm is presented respectively for internal and external datacenters to reduce the VM rental cost. To provide datacenter-level load balance and SLA (Service-Level Agreement) assurance, the proposed algorithm not only considers new arrival requests, but also re-dispatches requests being served to other datacenters. Finally, our work is evaluated and compared with the previous scheduling algorithms in small and large scale. Experimental results demonstrate the effectiveness of the proposed algorithm.
This work is supported by the National key research and development program of China (Grant No.2016YFB0201503, No.2016YFB0701101), National Natural Science Foundation of China (NSFC) (Grants No.61602205, No.51627805, No.61170004), Specialized Research Fund for the Doctoral Program of Higher Education (20130061110052), Major Special Research Project of Science and Technology Department of Jilin Province (20160203008GX), Key Science and Technology Research Project of Science and Technology Department of Jilin Province (20140204013GX). Jilin Scientific and Technological Development Program (20170520066JH), Graduate Innovation Fund of Jilin University.
- 1.Chen, J., Wang, C., Zhou, B.B., et al.: Tradeoffs between profit and customer satisfaction for service provisioning in the cloud. In: Proceedings of the 20th International Symposium on High Performance Distributed Computing, pp. 229–238. ACM, New York (2011)Google Scholar
- 2.Huang, D., Yang, D., Zhang, H., et al.: Energy-aware virtual machine placement in data centers. In: Global Communications Conference (GLOBECOM), pp. 3243–3249. IEEE, Anaheim (2012)Google Scholar
- 3.Anand, A., Lakshmi, J., Nandy, S.K.: Virtual machine placement optimization supporting performance SLAs. In: IEEE 5th International Conference on Cloud Computing Technology and Science (CloudCom), pp. 298–305. IEEE, Bristol (2013)Google Scholar
- 5.Popovici, F.I., Wilkes, J.: Profitable services in an uncertain world. In: Proceedings of the 2005 ACM/IEEE Conference on Supercomputing, p. 36. IEEE Computer Society, Seattle (2005)Google Scholar
- 10.Rao, L., Liu, X., Xie, L., et al.: Minimizing electricity cost: optimization of distributed internet data centers in a multi-electricity-market environment. In: 2010 IEEE INFOCOM Conference, pp. 1–9. IEEE, San Diego (2010)Google Scholar
- 11.Stanojevic, R., Shorten, R.: Distributed dynamic speed scaling. In: 2010 IEEE INFOCOM Conference, pp. 1–5. IEEE, San Diego (2010)Google Scholar
- 12.Goudarzi, H., Pedram, M.: Geographical load balancing for online service applications in distributed datacenters. In: IEEE CLOUD, pp. 351–358. IEEE, USA (2013)Google Scholar
- 13.Lin, M., Liu, Z., Wierman, A., et al.: Online algorithms for geographical load balancing. In: International Green Computing Conference (IGCC), pp. 1–10. IEEE, California (2012)Google Scholar
- 15.Qureshi, A.: Power-Demand Routing in Massive Geo-distributed Systems. Massachusetts Institute of Technology (2010)Google Scholar
- 16.Jing, C., Zhu, Y., Li, M.: Customer satisfaction-aware scheduling for utility maximization on geo-distributed cloud data centers. In: 2013 IEEE 10th International Conference on HPCC_EUC, pp. 218–225. IEEE, Zhangjiajie (2013)Google Scholar