Utility-maximizing server selection
Truong Khoa Phan, David P. Griffin, Elisa Maini, Miguel Rio · 2016
This paper presents a new method for selection between replicated servers distributed over a wide area, allowing application and network providers to trade-off costs with quality-of-service for their users. First, we create a novel utility framework that factors in quality of service metrics. Then we design a polynomial optimization algorithm to allocate user service requests to servers based on the utility while satisfying transit cost constraint. We then describe an efficient - low overhead distributed model with the need to only know a small subset of the data required by a global optimization formulation. Extensive simulations show that our method is scalable and leads to higher user utility compared with mapping user requests to the closest service replica.