QoS-Aware Matching of Edge Computing Services to Internet of Things
QoS-Aware Matching of Edge Computing Services to Internet of Things
复制标题
DOI：
10.1109/pccc.2018.8711148
复制
发表时间：
2018-11
期刊：
影响因子：
--
通讯作者：
Nafiseh Sharghivand;F. Derakhshan;Lena Mashayekhy
中科院分区：
文献类型：
--
作者：
Nafiseh Sharghivand;F. Derakhshan;Lena Mashayekhy
边缘计算是一种新的计算范式，旨在通过使计算资源更接近物联网（IoT）产生数据的地方来增强用户体验。 Cloudlet 是用户附近的附加基础设施组件，可促进边缘服务减少延迟和网络流量。物联网用户需要其应用程序的边缘服务满足严格的服务质量 (QoS)。一个关键的挑战是如何有效地将云与物联网应用相匹配，以实现对边缘计算服务的便捷随时访问。在本文中，我们通过考虑服务响应时间方面的 QoS 要求，提出新颖的边缘服务两侧匹配解决方案来解决这个问题。匹配机制提高了用户的体验质量。此外，我们还根据小云、物联网用户和系统的偏好和激励来确定边缘服务的动态定价。所提出的匹配是帕累托有效的、激励相容的、弱预算平衡的和计算高效的。我们通过广泛的性能分析实验进行全面评估，以评估我们提出的匹配和定价解决方案。
Edge computing is a new paradigm of computing, which aims at enhancing user experience by bringing computing resources closer to where data is produced by Internet of Things (IoT). Cloudlets, additional infrastructure components nearby users, facilitate edge services to decrease latency and network traffic. IoT users require edge services for their applications meeting a strict quality of service (QoS). A key challenge is how to efficiently match cloudlets to IoT applications to enable a convenient any-time access to edge computing services. In this paper, we address this problem by proposing novel two-sided matching solutions for edge services considering QoS requirements in terms of service response time. The matching mechanisms enhance the quality of experience of the users. In addition, we determine dynamic pricing of the edge services based on preferences and incentives of cloudlets, IoT users, and the system. The proposed matchings are Pareto-efficient, incentive compatible, weakly budget balanced, and computationally efficient. We perform a comprehensive assessment through extensive performance analysis experiments to evaluate our proposed matching and pricing solutions.