Pricing the spare bandwidth: towards maximizing data center’s profit

To provide tenants with predictable network performance, tenants are allowed to purchase specified amount of data center network (DCN) resources (e.g., bandwidth). Then, data center will reserve the purchased resources for the corresponding tenants. Meanwhile, some DCNs are work conserving, which means that the spare bandwidth will be fairly shared by all active tenants. Even though the amount of the spare bandwidth is stochastic and uncertain, DCN providers are not likely to give away it for free. Thus, data center with work conserving may impose extra payment on tenants for using the spare bandwidth. In this paper, we propose a suitable tariff to charge for the usages of DCN resources, which includes a bill for the usage of the spare bandwidth. Through theoretical analysis and simulation, we demonstrate that our tariff can incentivize tenants to adjust their purchases of bandwidths, which can lead to the improvement of data center’s profit by 27.4% without impairing the social welfare.摘要创新点1.提出了一种新型的数据中心网络用户效能建模方式。 这种建模方式不仅考虑到了用户数据发送速率, 也考虑到了用户带宽需求被完全满足的概率。2.分析了在可获得的剩余带宽为随机变量的情况下, 数据中心网络用户最优的预留带宽购买量。 我们分别展示了数据中心网络用户最优的预留带宽购买量和效用因子, 可获得的剩余带宽量, 以及对带宽计费方式的关系。3.通过理论分析和仿真测试, 我们证明我们提出的计费机制能够在不损失社会福利的前提下, 有效地提高数据中心网络的利润。

[1]  Chita R. Das,et al.  Characterizing Network Traffic in a Cluster-based, Multi-tier Data Center , 2007, 27th International Conference on Distributed Computing Systems (ICDCS '07).

[2]  Ning Ding,et al.  The only constant is change: incorporating time-varying network reservations in data centers , 2012, SIGCOMM.

[3]  R. Srikant,et al.  The Price of Simplicity , 2007, 2007 Conference Record of the Forty-First Asilomar Conference on Signals, Systems and Computers.

[4]  Sangtae Ha,et al.  A survey of smart data pricing , 2012, ACM Comput. Surv..

[5]  Du Xu,et al.  DistributedNet: A reasonable pricing and flexible network architecture for datacenter , 2014, 2014 IEEE International Conference on Communications (ICC).

[6]  Costas Courcoubetis,et al.  Pricing communication networks - economics, technology and modelling , 2003, Wiley-Interscience series in systems and optimization.

[7]  Michael Schwind,et al.  Dynamic Pricing and Automated Resource Allocation for Complex Information Services: Reinforcement Learning and Combinatorial Auctions , 2007 .

[8]  Baochun Li,et al.  Pricing cloud bandwidth reservations under demand uncertainty , 2012, SIGMETRICS '12.

[9]  Lucian Popa,et al.  What we talk about when we talk about cloud network performance , 2012, CCRV.

[10]  Sujata Banerjee,et al.  ElasticSwitch: practical work-conserving bandwidth guarantees for cloud computing , 2013, SIGCOMM.

[11]  Costas Courcoubetis,et al.  Pricing Communication Networks , 2003 .

[12]  A. Robert Calderbank,et al.  Layering as Optimization Decomposition: A Mathematical Theory of Network Architectures , 2007, Proceedings of the IEEE.

[13]  Di Xie,et al.  The only constant is change: incorporating time-varying network reservations in data centers , 2012, CCRV.

[14]  Dorgival O. Guedes,et al.  Gatekeeper: Supporting Bandwidth Guarantees for Multi-tenant Datacenter Networks , 2011, WIOV.

[15]  David A. Maltz,et al.  Network traffic characteristics of data centers in the wild , 2010, IMC '10.

[16]  Randy H. Katz,et al.  A view of cloud computing , 2010, CACM.

[17]  Jorge-Arnulfo Quiané-Ruiz,et al.  Runtime measurements in the cloud , 2010, Proc. VLDB Endow..