Edge server placement considering resilience in mobile edge computing networks
| dc.contributor.author | Begum, Syeda Mahfuza | |
| dc.contributor.supervisor | Pan, Jianping | |
| dc.date.accessioned | 2024-10-22T20:21:55Z | |
| dc.date.available | 2024-10-22T20:21:55Z | |
| dc.date.issued | 2024 | |
| dc.degree.department | Department of Computer Science | |
| dc.degree.level | Master of Science MSc | |
| dc.description.abstract | In today’s rapidly evolving communication landscape, the demand for exceptional Quality of Service (QoS) and Quality of Experience (QoE) in communication networks has reached unprecedented levels. This surge in demand can be attributed to the explosive growth and pervasive deployment of Internet infrastructure. Emerging technologies and novel applications underscore the urgency for a network architecture that not only delivers speed and efficiency but also boasts scalability and resilience beyond the capabilities of traditional cloud computing networks. Mobile Edge Computing (MEC) stands as a promising solution to address these challenges. By deploying Edge Servers (ESs) in close proximity to end-user devices, MEC enables the offloading of delay-sensitive and computationally intensive workloads from mobile applications. This deployment, in turn, mitigates latency issues and enhances the QoE for mobile users. However, the reliability of Edge Server Placement (ESP) within MEC networks is of paramount importance. While several studies have explored the ESP problem in MEC networks, they often focus on two main objectives: minimizing Edge Server (ES) access delay and optimizing workload distribution. However, one critical aspect has been relatively under-emphasized: the resiliency of ESP. The failure or malfunction of ESs, stemming from various challenges, can disrupt operations and degrade the overall QoS/QoE of the network. In this study, we tackle the ESP problem in MEC networks from a distinctive perspective. Our focal point is to minimize ES access delay, efficiently balance workloads, and significantly enhance network resilience. To achieve these objectives, our innovative algorithm employs a dual strategy. First, we utilize the robust K-medoids clustering algorithm for ESP, optimizing the architectural layout of MEC networks. Second, we introduce a bespoke heuristic algorithm designed to allocate multiple ESs to each Base Station (BS), thereby fortifying network resilience. This approach not only adheres to various constraints but also ensures uninterrupted services, even in the face of server failures, while consistently meeting key performance indicators. Experimental results, based on real-world data, prove the effectiveness of our algorithm. It not only reduces access delay and workload imbalances but also ensures responsive performance and uninterrupted services, even in scenarios involving ES failures. | |
| dc.description.scholarlevel | Graduate | |
| dc.identifier.uri | https://hdl.handle.net/1828/20621 | |
| dc.language | English | eng |
| dc.language.iso | en | |
| dc.rights | Available to the World Wide Web | |
| dc.subject | Mobile edge computing | |
| dc.subject | Edge server placement | |
| dc.subject | Network resilience | |
| dc.subject | Workload balancing | |
| dc.subject | Access delay | |
| dc.subject | Clustering | |
| dc.subject | Heuristic algorithm | |
| dc.subject | Edge server | |
| dc.subject | Base station | |
| dc.title | Edge server placement considering resilience in mobile edge computing networks | |
| dc.type | Thesis |