iDaaS: Inter-Datacenter Network as a Service

Abstract—Current-generation Internet-scale applications, such as video streaming, incur a large amount of wide area traffic. Trans-mitted on the unreliable Internet with no bandwidth guarantees, such traffic is suffering unpredictable network performance, which is however unappealing to the application providers. Fortunately, Internet giants like Google and Microsoft are increasingly deploying their private wide area networks (WANs) to connect their datacenters on the globe. Such private WANs are highly reliable, and can provide predictable network performance. In this paper, we propose a new type of service—inter-datacenter network as a service (iDaaS), where application providers, make bandwidth reservations for bandwidth guarantees from those Internet giants to support their wide area traffic. We study a bandwidth trading market of multiple iDaaS providers and application providers, and concentrate on the essential bandwidth pricing problem. The challenging issue involves in pricing in such a market is that the benefits of both iDaaS providers and application providers are close-knit connected. To address this issue, we model the interaction between iDaaS providers and application providers as a Stackelberg game, and analyze the existence and uniqueness of the equilibrium. We further present an efficient bandwidth price computation algorithm by blending the advantage of a geometrical Nash bargaining solution and the demand segmentation method. Based on the computed price, we present two bandwidth reservation algorithms, where each iDaaS provider’s bandwidth are reserved in a weighted fair manner and a max-min fair manner, respectively. Finally, we conduct comprehensive trace-driven experiments. The evaluation results show the efficiency of our proposed algorithms in the bandwidth market. F