Achieving Energy Efficiency in Cloud Brokering

The proliferation of cloud providers has brought substantial interoperability complexity to the public cloud market, in which cloud brokering has been playing an important role. However, energy-related issues for public clouds have not been well addressed in the literature. In this paper, we claim that the broker is also situated in a perfect position where necessary actions can be taken to achieve energy efficiency for public cloud systems, particularly through job assignment and scheduling. We formulate the problem by a mixed integer program and prove its NP-hardness. Based on the complexity analysis, we simplify the problem by introducing admission control on jobs. In the sequel, optimal job assignment can be done straightforwardly and the problem is transformed into improving job admission rate by scheduling on two coupled phases: data transfer and job execution. The two scheduling phases are further decoupled and we develop efficient scheduling algorithm for each of them. Experimental results show that the proposed solution can achieve significant reduction on energy consumption with admission rates improved as well, even in large-scale public cloud systems.

[1]  Christos Koulamas,et al.  Flexible flow shop scheduling with uniform parallel machines , 2006, Eur. J. Oper. Res..

[2]  Pascal Bouvry,et al.  Cloud Brokering: Current Practices and Upcoming Challenges , 2015, IEEE Cloud Computing.

[3]  Lachlan L. H. Andrew,et al.  Dynamic Right-Sizing for Power-Proportional Data Centers , 2011, IEEE/ACM Transactions on Networking.

[4]  Paramvir Bahl,et al.  Low Latency Geo-distributed Data Analytics , 2015, SIGCOMM.

[5]  Walter Cerroni Performance of network and computing resource sharing in federated cloud systems , 2014, DCC '14.

[6]  Carlo Curino,et al.  Global Analytics in the Face of Bandwidth and Regulatory Constraints , 2015, NSDI.

[7]  Shaolei Ren,et al.  Paying to save: Reducing cost of colocation data center via rewards , 2015, 2015 IEEE 21st International Symposium on High Performance Computer Architecture (HPCA).

[8]  Haiying Shen,et al.  Harnessing the Power of Multiple Cloud Service Providers: An Economical and SLA-Guaranteed Cloud Storage Service , 2015, 2015 IEEE 35th International Conference on Distributed Computing Systems.

[9]  David S. Johnson,et al.  Computers and Intractability: A Guide to the Theory of NP-Completeness , 1978 .

[10]  Baochun Li,et al.  Temperature Aware Workload Managementin Geo-Distributed Data Centers , 2013, IEEE Trans. Parallel Distributed Syst..

[11]  Athanasios V. Vasilakos,et al.  GreenDCN: A General Framework for Achieving Energy Efficiency in Data Center Networks , 2013, IEEE Journal on Selected Areas in Communications.

[12]  Haiying Shen,et al.  Towards green cloud computing: Demand allocation and pricing policies for cloud service brokerage , 2015, IEEE BigData.

[13]  Shaolei Ren,et al.  Provably-Efficient Job Scheduling for Energy and Fairness in Geographically Distributed Data Centers , 2012, 2012 IEEE 32nd International Conference on Distributed Computing Systems.