Efficient algorithms to minimize the end-to-end latency of edge network function virtualization
- PDF / 1,889,096 Bytes
- 12 Pages / 595.276 x 790.866 pts Page_size
- 110 Downloads / 183 Views
ORIGINAL RESEARCH
Efficient algorithms to minimize the end‑to‑end latency of edge network function virtualization Karanbir Singh Ghai1 · Salimur Choudhury1 · Abdulsalam Yassine2 Received: 24 July 2019 / Accepted: 30 October 2019 © Springer-Verlag GmbH Germany, part of Springer Nature 2020
Abstract In future wireless networks, network function virtualization will lay the foundation for establishing a new dynamic resource management framework to efficiently utilize network resources. The main problem discussed in this paper is the minimization of total latency for an edge network and how to solve it efficiently. A model of users, virtual network functions and hosting devices has been taken, and is used to find the minimum latency using integer linear programming. The problem is NP-hard and takes exponential time to return the optimal solution. We apply the stable matching based algorithm to solve the problem in polynomial time and then utilize local search to improve its efficiency further. From extensive performance evaluation, it is found that our proposed algorithm is very close to the optimal scheme in terms of latency and better in terms of time complexity. Keywords Network function virtualization · Hosting device · Latency · Stable matching · Local search
1 Introduction In today’s time, we require an efficient and advanced network model that can support the growing load of the users (Hu et al. 2011). Different models used for network computing are centralized network computing and distributed network computing. In the initial phases of networking the centralized network model was used as there were not many devices that could support the whole networks but trend changed and we now use more of distributed networking model. The main reason behind this is, in centralized networks the complete load of the network system falls on one central machine which increases the risk of network failure but in distributed networks, the network relies on various nodes or network devices which makes it more efficient and * Karanbir Singh Ghai [email protected] Salimur Choudhury [email protected] Abdulsalam Yassine [email protected] 1
Department of Computer Science, Lakehead University, Thunder Bay, ON, Canada
Department of Software Engineering, Lakehead University, Thunder Bay, ON, Canada
2
thus more reliable (Baran 1964). Edge Networks as the name suggests is a distributed computing paradigm in which computation is wholly or mostly performed on distributed device nodes known as smart devices or edge devices as opposed to primarily taking place in a centralized cloud environment. Here “edge” is defined as any computing and network resources along the path between data sources and cloud data centers (Shi et al. 2016). For example, a smart phone is the edge between smart body sensors and a cloud, a gateway in a smart home is the edge between smart home things and a cloud. Edge computing is related to the concepts of wireless sensor networks, intelligent and context-aware networks and smart objects
Data Loading...
