Now You See Me: Embedded Mobile Edge Computing Awareness for Optimizing Multi-Task Applications in Constrained IoT Devices
Jaime Sebastian Burbano, Francisco Salgado, Rolf Schuster · 2024
Mobile Edge Computing (MEC) places computing and storage resources on the network's edge, which opens the door for implementing an entirely new set of delay-sensitive and compute-intensive applications in constrained IoT devices. Nevertheless, from the perspective of an IoT device, using MEC introduces several challenges that need to be addressed. One is the proper execution and resource utilization of device-native and edge-dependent tasks, especially when considering multi-task edge applications. Current real-time schedulers do not contemplate factors inherent to the nature of MEC, such as shared resource utilization, dynamic execution site migrations, the influence of remote computing, etc. Therefore, this paper introduces an agent embedded in IoT devices to provide real-time schedulers with mobile edge computing awareness. Moreover, the performance of the proposed solution is assessed by running two multi-task edge applications in the smart vehicles domain, depicting a considerable task throughput enhancement when compared to existing real-time schedulers.