A Device-Edge-Cloud Collaborative Framework for Hierarchical Computation Offloading

With the ever increasing demands of resource-hungry applications, computational tasks tend to be offloaded to edge servers for execution. However, most of the existing works on edge computing integrated with cloud computing ignore the additional computational resources of end devices and neighboring edge servers. Hierarchical task offloading allows more flexible distribution of computation loads among devices, edges, and the cloud hence further improving the quality of the computation experience. In this paper, a device-edge-cloud collaboration framework (DECHO) for hierarchical task offloading is developed, which learns to explore all available resources to improve offloading performance in a decentralized manner. Furthermore, an experimental prototype of DECHO is demonstrated. The experimental results present the task execution efficiency performance, which validates the scalability and effectiveness of the proposed framework.