Server deployment strategies considering robustness and resource constraints in edge computing

Abstract

Abstract Edge computing, as an emerging computing paradigm, efficiently offloads computationally intensive tasks to edge servers, extending services from the cloud center to the edge and significantly improving the efficiency of network data processing. However, existing research mostly focuses on the ideal deployment and optimization of edge servers, such as minimizing latency and load balancing, with little attention given to strategies for handling abnormal situations like soft attacks or sudden failures. This study delves into how to formulate server deployment strategies aiming to maximize system robustness and minimize costs when edge servers face soft attacks or sudden failures. Drawing insights from graph theory and network connectivity, we propose a server deployment strategy that enhances system robustness under resource constraints. Specifically, we introduce the concept of edge-delay-tolerant networks, ensuring the rapid establishment of a complete “backup link” in case of service interruption by constructing backup relay nodes and routing information tables to safeguard system continuity and stability. Additionally, we innovatively design adaptive edge server deployment methods in “silent” and “active” modes to cater to varying demands in different fault scenarios. To validate the effectiveness of the proposed strategies, we construct a multi-objective optimization problem through mathematical modeling and devise an hybrid optimization algorithm for solving it. Experimental simulation results demonstrate that the algorithm can provide near-optimal performance, effectively enhancing system robustness under resource constraints. This research not only offers new insights and methods for fault response mechanisms in edge computing but also provides valuable references for practical server deployment strategies, thereby promoting further development of edge computing technology.