READ: Robustness-Oriented Edge Application Deployment in Edge Computing Environment
Bo Li, Qiang He, Guangming Cui, Xiaoyu Xia, Feifei Chen, Hai Jin, Yun Yang
Abstract
In recent years, edge computing has emerged as a prospective distributed computing paradigm that overcomes several limitations of cloud computing. In the edge computing environment, a service provider can deploy its application instances on edge servers at the edge of the network to serve its own users with low latency. Given a limited budget <inline-formula><tex-math notation="LaTeX">$\mathcal {K}$</tex-math></inline-formula> for deploying applications on the edge servers in a particular geographical area, a number of approaches have been proposed very recently to determine the optimal deployment strategy that achieves various optimization objectives, e.g., to maximize the servers’ coverage, to minimize the average network latency, etc. However, the robustness of the services collectively delivered by the service provider’s applications deployed on the edge servers has not been considered at all. This is a critical issue, especially in the highly distributed, dynamic and volatile edge computing environment. In this article, we make the first attempt to tackle this challenge. Specifically, we formulate this <i>Robustness-oriented Edge Application Deployment</i> (READ) problem as a constrained optimization problem and prove its <inline-formula><tex-math notation="LaTeX">$\mathcal {NP}$</tex-math></inline-formula> -hardness. Then, we provide an integer programming based approach named READ- <inline-formula><tex-math notation="LaTeX">$\mathcal {O}$</tex-math></inline-formula> for solving this problem precisely. We also provide an approximation algorithm, namely READ- <inline-formula><tex-math notation="LaTeX">$\mathcal {A}$</tex-math></inline-formula> , for finding near-optimal solutions to large-scale READ problems efficiently. We prove its approximation ratio is not worse than <inline-formula><tex-math notation="LaTeX">$\mathcal {K}/2$</tex-math></inline-formula> , which is a constant regardless of the total number of edge servers. We evaluate our approaches experimentally on a widely-used real-world dataset against five representative approaches. The experiment results demonstrate that our approaches can solve the READ problem effectively and efficiently.