An Envy-Free Auction Mechanism for Resource Allocation in Edge Computing Systems

摘要

One of the major challenges in Mobile Edge Computing~(MEC) systems is to decide how to allocate and price edge/cloud resources so that a given system's objective, such as revenue or social welfare, is optimized. One promising approach is to allocate these resources based on auction models, in which users place bids for using a certain amount of resources. In this paper, we address the problem of resource allocation and pricing in a two-level edge computing system. We consider a system in which servers with different capacities are located in the cloud or at the edge of the network. Mobile users compete for these resources and have heterogeneous demands. We design an auction-based mechanism that allocates and prices edge/cloud resources. The proposed mechanism is novel in the sense that it handles the allocation of resources available at the two-levels of the system by combining features from both position and combinatorial auctions. We show that the proposed mechanism is individually-rational and produces envy-free allocations. The first property guarantees that users are willing to participate in the mechanism, while the second guarantees that when the auction is finished, no user would be happier with the outcome of another user. We evaluate the performance of the proposed mechanism by performing extensive experiments. The experimental results show that the proposed mechanism is scalable and obtains efficient solutions.