ApMove: A service migration technique for connected and autonomous vehicles

Zakarya, Muhammad, Gillam, Lee, Khan, Ayaz Ali, Rana, Omer ORCID: https://orcid.org/0000-0003-3597-2646 and Buyya, Rajkumar 2024. ApMove: A service migration technique for connected and autonomous vehicles. IEEE Internet of Things Journal 11 (17) , pp. 28721-28733. 10.1109/JIOT.2024.3403415

Preview

PDF - Accepted Post-Print Version Download (453kB) | Preview

Abstract

Multi-access edge computing systems (MECs) bring the capabilities of cloud computing closer to the radio access network (RAN), in the context of 4G and 5G telecommunication systems, and converge with existing radio access technologies like satellite or WiFi. An MEC is a cloud server that runs at the mobile network’s edge and is installed and executed using virtual machines (VMs), containers, and/or functions. A cloudlet is similar to an MEC that consists of many servers which provide real-time, low-latency, computing services to connected users in close proximity. In connected vehicles, services may be provisioned from the cloud or edge that will be running users’ applications. As a result, when users travel across many MECs, it will be necessary to transfer their applications in a transparent manner so that performance and connectivity are not negatively affected. In this paper, we propose an effective strategy for migrating connected users’ services from one edge to another or, more likely, to a remote cloud in an MEC. A mathematical model is presented to estimate the expected times to allocate and migrate services. Our evaluations, based on real workload traces and mobility patterns, suggest that the proposed strategy “ApMove" migrates connected services while ensuring their performance ( 0.004% – 2.99% loss), reduced runtimes, therefore, users’ costs ( 4.3% – 11.63%), and minimizing the response time ( 7.45% – 9.04%). Furthermore, approximately 17.39% migrations are avoided. We also study the impacts of variations in the car’s speed and network transfer rates on service migration durations, latencies, and service execution times.

Item Type:	Article
Date Type:	Publication
Status:	Published
Schools:	Computer Science & Informatics
Publisher:	Institute of Electrical and Electronics Engineers
ISSN:	2327-4662
Funders:	UKRI
Date of First Compliant Deposit:	29 May 2024
Date of Acceptance:	16 May 2024
Last Modified:	04 Sep 2024 15:53
URI:	https://orca.cardiff.ac.uk/id/eprint/169069

Actions (repository staff only)

Edit Item