Energy-Efficient Offloading for DNN-Based Smart IoT Systems in Cloud-Edge Environments

Deep Neural Networks (DNNs) have become an essential and important supporting technology for smart Internet-of-Things (IoT) systems. Due to the high computational costs of large-scale DNNs, it might be infeasible directly deploy them in energy-constrained IoT devices. Through offloading computation-intensive tasks cloud or edges, computation offers a feasible solution execute DNNs. However, energy-efficient DNN based systems with deadline constraints cloud-edge environments is still open challenge. To address this challenge, we first design new system energy consumption model, which takes into account runtime, switching, computing all participating servers (from both edge) Next, novel strategy on Self-adaptive Particle Swarm Optimization algorithm using Genetic Algorithm operators (SPSO-GA) proposed. This can efficiently make decisions layers layer partition operations, lessen encoding dimension improve execution time SPSO-GA. Simulation results demonstrate that proposed significantly reduce compared other classic methods.