Edge Pricing Mechanisms under Cloud Tiered Pricing

As a supplement to cloud computing, edge computing provides low-latency computing services by deploying edge servers on the user side. Therefore, users are more inclined to deploy tasks on edge platforms; however, there are few studies on the pricing mechanism of edge platforms. In this paper, we propose a model to solve the Edge Price under Cloud Tiered Pricing problem, which edge platforms can design ideal pricing strategy to maximize their own revenue under the tiered pricing strategy of cloud platforms. In order to design an ideal and truthful mechanism to solve the problem of the model, we successively propose three different truthful mechanism algorithms: greedy, dynamic programming and random sampling. Extensive simulations were subsequently performed on the Google Cluster dataset. From the experimental results, we find that the dynamic programming mechanism can produce results as similar as the optimal results in pseudo-polynomial time. With the increase of user requests, the greedy mechanism can obtain better results under the comprehensive consideration of running time and total results. When the total requests of users are less than the capacity of the edge platform, the random sampling mechanism is a truthful algorithm and the performance can also be guaranteed.