Decentralized Edge Caching under Budget and Storage Constraints: A Game-Theoretic Approach

The rapid growth of mobile social networks (MSNs) has driven demand for low-latency content delivery, pushing edge caching to the forefront. In practice, multiple content providers (CPs) compete for the finite storage on edge devices (EDs), each facing unique budget and operational constraints. Hamta Sedghani, Zahra Seyedi, Mauro Passacantando, and Danilo Ardagna propose a hierarchical game-theoretic framework that captures both the strategic interplay between CPs and EDs (via Stackelberg) and the non-cooperative game among CPs themselves. This dual-layer approach accounts for real-world budget limits and storage scarcity, aiming to find stable, efficient resource allocations without central coordination.

The key insight: under light storage constraints, the CP competition constitutes an exact potential game, which guarantees the existence of a pure-strategy Nash equilibrium and enables decentralized convergence using simple learning dynamics. When storage becomes binding, the game loses this structure, yet extensive simulations demonstrate that convergence remains stable and efficient in practice. The analysis further reveals that storage scarcity fundamentally alters economic outcomes—amplifying inequality among CPs while increasing the relative bargaining power of edge devices. These findings offer a scalable, economically grounded blueprint for decentralized resource allocation in emerging multi-provider edge caching systems.