Fault-Tolerating Edge Computing with Server Redundancy Based on a Variant of Group Degree Centrality
In the distributed and dynamic edge computing environment, edge servers are subject to runtime failures. Therefore, edge servers in an area must be fault-tolerated to ensure the reliability of services deployed on those edge servers. Server redundancy is
- PDF / 1,530,130 Bytes
- 17 Pages / 439.37 x 666.142 pts Page_size
- 42 Downloads / 158 Views
2
1 Wuhan University of Technology, Wuhan, China {whutduwei,zhangxiran,wliu,saltfish,284375,yangyc}@whut.edu.cn Hubei Key Laboratory of Transportation Internet of Things, Wuhan, China 3 Swinburne University of Technology, Hawthorn, Australia {qhe,gcui}@swin.edu.au 4 Deakin University, Burwood, Australia [email protected]
Abstract. In the distributed and dynamic edge computing environment, edge servers are subject to runtime failures. Therefore, edge servers in an area must be fault-tolerated to ensure the reliability of services deployed on those edge servers. Server redundancy is an effective fault tolerance technique and has been widely applied in different distributed computing environments in the past decade. However, conventional fault tolerance techniques are not suitable for edge computing which has unique characteristics, i.e., the constrained coverage areas of individual edge servers (coverage constraint) and the partial overlapping between edge servers’ coverage areas (overlapping constraint). In this paper, we make the first attempt to investigate and tackle the novel edge server redundancy (ESR) problem. We prove that the ESR problem is N Phard. Then, we introduce a novel optimal approach for identifying a group of edge servers to be redundant. The objective is to maximize the effectiveness of fault tolerance measured by the harmonic mean of the scope and strength of fault tolerance given a redundancy budget. Furthermore, we propose a heuristic approach for finding sub-optimal fault tolerance strategies efficiently in large-scale ESR scenarios. Extensive experiments are conducted on a widely-used real-world dataset to evaluate the proposed approaches against three representative baseline approaches. Keywords: Edge computing · Fault tolerance identification · Group degree centrality
1
· Redundant server
Introduction
Edge computing is a promising distributed computing paradigm that provides computation and storage capacity within end-users’ proximity [10]. In such an c Springer Nature Switzerland AG 2020 E. Kafeza et al. (Eds.): ICSOC 2020, LNCS 12571, pp. 198–214, 2020. https://doi.org/10.1007/978-3-030-65310-1_16
Fault-Tolerating Edge Computing with Server Redundancy
199
environment, edge servers, typically facilitated by a micro-data center or a small cluster of servers, usually bind with small base stations to share the computation burdens of mobile and IoT devices and provide high quality-of-service (QoS) for end-users [16]. Therefore, edge servers are the critical components in the edge computing environment. However, edge servers are subject to runtime failures because of various unexpected reasons such as hardware faults, software exceptions, and cyberattacks [3], similar to their counterparts in cloud data centers [22]. In fact, edge servers are more prone to failures and outages than cloud servers due to their geographical dispersion, limited resources and low scalability [1]. In addition, unlike cloud servers that are managed in-house, geographicallydistributed edge servers cannot b
Data Loading...
