A Mixed-Integer Bi-level Model for Joint Optimal Edge Resource Pricing and Provisioning

TL;DR

This paper introduces a bi-level optimization model for joint edge resource pricing and provisioning, addressing the complex interplay between platform profit maximization and service cost minimization in edge computing.

Contribution

It develops a novel decomposition-based algorithm to solve the challenging bi-level mixed-integer problem exactly, which was previously intractable with traditional methods.

Findings

The proposed algorithm efficiently finds exact solutions.

Numerical results validate the model's effectiveness.

The approach outperforms existing heuristics.

Abstract

This paper studies the joint optimization of edge node activation and resource pricing in edge computing, where an edge computing platform provides heterogeneous resources to accommodate multiple services with diverse preferences. We cast this problem as a bi-level program, with the platform acting as the leader and the services as the followers. The platform aims to maximize net profit by optimizing edge resource prices and edge node activation, with the services' optimization problems acting as constraints. Based on the platform's decisions, each service aims to minimize its costs and enhance user experience through optimal service placement and resource procurement decisions. The presence of integer variables in both the upper and lower-level problems renders this problem particularly challenging. Traditional techniques for transforming bi-level problems into single-level formulations are inappropriate owing to the non-convex nature of the follower problems. Drawing inspiration from the column-and-constraint generation method in robust optimization, we develop an efficient decomposition-based iterative algorithm to compute an exact optimal solution to the formulated bi-level problem. Extensive numerical results are presented to demonstrate the efficacy of the proposed model and technique.