Container-based Task Offloading for Time-Critical Fog Computing

Abstract

Moving applications from the local Internet of Things (IoT) device to the neighboring infrastructures has been the highlight of the current distributed computing era. Due to high latency and network bottlenecks, Cloud Computing has become rather problematic for real-time applications. Accordingly, Fog Computing, well supported by 5G, has been presented as the missing piece to provide along with cloud computing for virtually all applications. Application virtualization through containers has become more and more popular in the context of Cloud Computing as it simplifies live migration of the application or it’s selected tasks. The concept has been adapted to Fog computing too, However, time-critical task migration and in particular task offloading have not been sufficiently investigated. This work proposes a novel approach to offload time-critical application tasks from the application initiator to surrounding Fog nodes based on Docker containers and Checkpointing. The result is a flexible layer-oriented framework that effectively restricts the offloading to only necessary steps in only two message rounds. Performance evaluations show that through careful easy arrangements the framework performs remarkably well concerning offloading latency.