Research Article
Stochastic Performance Analysis of Edge-Device Collaboration
@INPROCEEDINGS{10.1007/978-3-030-89814-4_7, author={Zhidu Li and Xue Jiang and Ruyan Wang and Dapeng Wu}, title={Stochastic Performance Analysis of Edge-Device Collaboration}, proceedings={Mobile Multimedia Communications. 14th EAI International Conference, Mobimedia 2021, Virtual Event, July 23-25, 2021, Proceedings}, proceedings_a={MOBIMEDIA}, year={2021}, month={11}, keywords={Edge-device collaboration Stochastic delay guarantee Task offloading Resource allocation}, doi={10.1007/978-3-030-89814-4_7} }- Zhidu Li
Xue Jiang
Ruyan Wang
Dapeng Wu
Year: 2021
Stochastic Performance Analysis of Edge-Device Collaboration
MOBIMEDIA
Springer
DOI: 10.1007/978-3-030-89814-4_7
Abstract
As a new computing paradigm, edge-device collaboration is able to reduce the task execution delay for the devices with limited energy. However, deterministic delay guarantee is unavailable due to the random task arrivals and time-varying channel fading. This paper proposes to study the task execution delay of edge-device collaboration from the probabilistic view of point. Specifically, an optimization problem is formulated with objective of minimizing the proportion of tasks that are not served in time. Then, a general delay violation probability for stochastic system is derived, based on which, the task execution delay violation probability at local device and that at the edge server are obtained respectively. Thereafter, the optimization problem can be transferred into a tractable one where task offloading proportion and task offloading rate can be jointly optimized. The effectiveness of the proposed scheme is finally validated by extensive numerical simulations.
