PROJECT TITLE :

A Shapley-Value Mechanism for Bandwidth On Demand between Datacenters - 2018

ABSTRACT:

Recent studies in cloud resource allocation and pricing have centered on computing and storage resources however not network bandwidth. Cloud users today usually deploy services across multiple geo-distributed datacenters, with significant inter-datacenter traffic generated, paid by cloud providers to ISPs. An effective bandwidth allocation and charging mechanism is required between the cloud supplier and the cloud users. Existing volume based mostly static charging schemes lack market efficiency. This work presents the primary dynamic pricing mechanism for inter-knowledge-center on-demand bandwidth, via a Shapley worth primarily based auction. Our auction is expressive enough to simply accept bids as a flat bandwidth rate plus a time length, or a knowledge volume with a transfer deadline. We have a tendency to start with an offline auction, design an optimal end-to-finish traffic scheduling approach, and exploit the Shapley price in computing payments. Our auction is truthful, individual rational, budget balanced and approximately efficient in social welfare. An online version of the auction follows, where selections are created instantly upon the arrival of each user's realtime transmission demand. We tend to propose an economical on-line traffic scheduling algorithm, and approximate the offline Shapley worth based mostly payments on the fly. We have a tendency to validate our mechanism design with solid theoretical analysis, as well as trace-driven simulation studies.