A game-theoretic approach to coalition formation in fog provider federations

Abstract

In this paper we deal with the problem of making a set of Fog Infrastructure Providers (FIPs) increase their profits when allocating their resources to process the data generated by IoT applications that need to meet specific QoS targets in face of time-varying workloads. We show that if FIPs cooperate among them, by mutually sharing their workloads and resources, then each one of them can improve its net profit. By using a game-theoretic framework, we study the problem of forming stable coalitions among FIPs. Furthermore, we propose a mathematical optimization model to allocate IoT applications to a set of FIPs, in order to reduce costs and, at the same time, to meet the corresponding QoS targets. Based on this, we propose an algorithm, based on cooperative game theory, that enables each FIP to decide with whom to cooperate in order to increase its profits. The effectiveness of the proposed algorithm is demonstrated through an experimental evaluation considering various workload intensities. The results we obtain from these experiments show the ability of our algorithm to form coalitions of FIPs that are stable and profitable in all the scenarios we consider.

Profit‐aware coalition formation in fog computing providers: A game‐theoretic approach

“…Last but not the least, this paper extends our previous work 18,57 by introducing a different coalition formation allocation strategy that improves the stability of the formed coalitions, guarantees the efficiency of the payoff allocation, and ensures fairness in the payoff allocation thanks to the use of the Shapley value (note that a fair payoff allocation is a mandatory property to convince autonomous agents (often competing) to adopt the solution we propose).…”

“…We suppose that the load of data processing requests sent to application S i in any geographic area j changes with time and is described by its load profile curve i, j (t) expressing, as function of time t, the rate at which requests are submitted (eg, see other works 18,25,26 ). Moreover, we suppose that, for each application S i and area j, i, j (t) is known in advance.…”

“…In particular, we propose a game‐theoretic framework that extends our previous work 18 to study the problem of forming stable coalitions among FIPs, and a mathematical optimization model to allocate IoT applications to a group of resources, in order to increase profits and, at the same time, to meet application QoS for FIPs inside the same coalition. As a result, we devise a decision algorithm that, taking in input the capacity of fog resources of each FIP, the QoS levels negotiated with its clients, the intensity of the workload submitted to these resources, and the revenues and costs of running its own workloads and the workloads of other FIPs on its own these resources, allows it to assess the profitability of joining or not a coalition formed by other FIPs.…”

AI-Enhanced Load Balancing in Federated Edge Computing Networks: Challenges and Solutions

“…Especially, designing of new pricing models for the federation of resources among EC nodes and policies for resource sharing among EC nodes schemes appear to be highly essential. When a group of SPs are willing to cooperate among them- selves to maximize their utilities, then cooperative game-theoretic models can be proposed where a group of SPs form stable coalitions [6]. However, if the SPs are completely non-cooperative, we can propose load balancing mechanisms based on noncooperative game-theoretic models [7].…”

Profit Maximization by Forming Federations of Geo-Distributed MEC Platforms

“…We propose an approach to federation structure formation in the framework of coalition game. Coalition game theory has been applied to grid computing [6], vehicular cloud [7], mobile cloud [8], traditional cloud [3], and fog computing [9]. Our approach differs from prior work in the following points.…”