Edge Computing Efficiency and Sustainability

The field of edge computing is moving towards improving efficiency and sustainability. Researchers are exploring innovative solutions to reduce latency, energy consumption, and carbon emissions. One notable direction is the development of management frameworks that combine energy-aware task allocation with revenue-sharing mechanisms, enabling low-latency task execution and effective monetization of resources. Another area of focus is the design of dynamic and uncertain edge networks that facilitate efficient and timely computing services. Noteworthy papers include:

  • A Management Framework for Vehicular Cloud, which proposes a management scheme for VCC that combines energy-aware task allocation with a game-theoretic revenue-sharing mechanism, reducing CO2 emissions by more than 99% compared to conventional edge infrastructures.
  • Oh-Trust, which explores a dynamic and uncertain edge network and proposes an overbooking- and hybrid trading-empowered resource scheduling mechanism with reputation update, enhancing resource utilization and profitability for both parties.
  • PAST, which integrates two complementary mechanisms, PilotAO and AdaptAO, to enable stability and flexibility in edge-assisted UAV networks, guaranteeing individual rationality, strong stability, competitive equilibrium, and weak Pareto optimality.
  • Accuracy vs Performance, which presents a solution for low-latency deadline-constrained DNN offloading on mobile edge devices, designing a scheduling algorithm with lightweight network state representation and dynamic bandwidth estimation mechanism.

Sources

A Management Framework for Vehicular Cloudtoward Economic and Environmental Efficiency

Oh-Trust: Overbooking and Hybrid Trading-empowered Resource Scheduling with Smart Reputation Update over Dynamic Edge Networks

PAST: Pilot and Adaptive Orchestration for Timely and Resilient Service Delivery in Edge-Assisted UAV Networks under Spatio-Temporal Dynamics

Accuracy vs Performance: An abstraction model for deadline constrained offloading at the mobile-edge

Built with on top of