Mechanism Design for Time-Varying Value Tasks in High-Load Edge Computing Markets

A large number of computing task requests are generated by user terminals during peak hours in high-demand areas, but the resource capacity of edge servers is limited. It is necessary to design appropriate resource allocation and pricing mechanisms to address this resource competition dilemma. This paper proposes an auction-based mechanism called GMPO from an economic perspective. A market where multiple buyers and sellers compete with each other is considered, and the auction mechanisms is used to prevent these entities from falsely reporting information. As an extension of the concept of the age of information, the value of delay-sensitive computing tasks will decrease over time. This paper allocates resources greedily according to defined priorities and charge based on critical prices. The experiment results demonstrate that the proposed mechanism can effectively improve social welfare and guarantee the economic properties of auctions.