Equality or Efficiency: A Game of Distributed Multi-type Fair Resource Allocation on Computational Agents

We propose a game theoretical approach to multi-type fair resource allocation algorithms in a distributed heterogeneous environment. Distributed approaches can benefit large scale computing systems, especially if the hardware computing resources in the system are organized in a decentralized manner as in federated clusters and cyber physical systems. The simulation results show that our approach has comparable performance to well-known centralized and decentralized approaches in fairness and resource utilization, while relies less on communications.