论文标题
真实的计算卸载机制用于边缘计算
Truthful Computation Offloading Mechanisms for Edge Computing
论文作者
论文摘要
边缘计算（EC）是一个有希望的范式，为网络边缘的用户提供了分布式计算解决方案。将其计算为EC时，为用户保留令人满意的经验质量（QOE）是一个非平凡的问题。 EC中的计算卸载需要共同优化用户的访问点（AP）分配和边缘服务放置，这在计算上由于其组合性质而在计算上很棘手。此外，用户是自私的，他们可能会错误地报告其偏好，从而导致资源分配和网络拥塞效率低下。在本文中，我们解决了这个问题，并设计了一种基于算法机制设计的新型机制，以实现系统平衡。我们的机制分配了一对适当的AP和Edge服务器，以及每个新的加入用户的服务价格，使即时社交盈余最大化，同时满足EC系统中所有用户的偏好。宣布真正的偏好是用户的弱势策略。实验结果表明，就经验丰富的端到端潜伏期而言，我们的机制优于用户平衡和随机选择策略。
Edge computing (EC) is a promising paradigm providing a distributed computing solution for users at the edge of the network. Preserving satisfactory quality of experience (QoE) for users when offloading their computation to EC is a non-trivial problem. Computation offloading in EC requires jointly optimizing access points (APs) allocation and edge service placement for users, which is computationally intractable due to its combinatorial nature. Moreover, users are self-interested, and they can misreport their preferences leading to inefficient resource allocation and network congestion. In this paper, we tackle this problem and design a novel mechanism based on algorithmic mechanism design to implement a system equilibrium. Our mechanism assigns a proper pair of AP and edge server along with a service price for each new joining user maximizing the instant social surplus while satisfying all users' preferences in the EC system. Declaring true preferences is a weakly dominant strategy for the users. The experimental results show that our mechanism outperforms user equilibrium and random selection strategies in terms of the experienced end-to-end latency.
