A Multi–Source Fluid Queue Based Stochastic Model of the Probabilistic Offloading Strategy in a MEC System With Multiple Mobile Devices and a Single MEC Server

• Published in: International journal of applied mathematics and computer science Vol. 32; no. 1; pp. 125 - 138
• Main Authors: Zheng, Huan, Jin, Shunfu
• Format: Journal Article
• Language: English
• Published: Zielona Góra Sciendo 01.03.2022; De Gruyter Poland
• Subjects: birth-and-death process; Buffers; Computation offloading; cumulative distribution function; Distribution functions; Edge computing; Electronic devices; Mobile computing; mobile edge computing; multi-source fluid queue; Network latency; Performance evaluation; probabilistic offloading strategy; Probability theory; Queues; Response time; Servers; Stochastic models
• ISSN: 1641-876X; 2083-8492
• DOI: 10.34768/amcs-2022-0010

Mobile edge computing (MEC) is one of the key technologies to achieve high bandwidth, low latency and reliable service in fifth generation (5G) networks. In order to better evaluate the performance of the probabilistic offloading strategy in a MEC system, we give a modeling method to capture the stochastic behavior of tasks based on a multi-source fluid queue. Considering multiple mobile devices (MDs) in a MEC system, we build a multi-source fluid queue to model the tasks offloaded to the MEC server. We give an approach to analyze the fluid queue driven by multiple independent heterogeneous finite-state birth-and-death processes (BDPs) and present the cumulative distribution function (CDF) of the edge buffer content. Then, we evaluate the performance measures in terms of the utilization of the MEC server, the expected edge buffer content and the average response time of a task. Finally, we provide numerical results with some analysis to illustrate the feasibility of the stochastic model built in this paper.