当前位置: X-MOL 学术Future Gener. Comput. Syst. › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
Multi-round auction-based resource allocation for edge computing: Maximizing social welfare
Future Generation Computer Systems ( IF 7.5 ) Pub Date : 2022-11-04 , DOI: 10.1016/j.future.2022.10.028
Jianmao Xiao, Qinghang Gao, Zhenyue Yang, Yuanlong Cao, Hao Wang, Zhiyong Feng

5G NR-Lite technology still suffers from inadequate response to low latency, high speed and mass connectivity scenarios in high mobility conditions. Building on 4G and 5G, 6G networks will work to more fully enable the Internet of Things (IoT) paradigm, working to provide secure wireless computing for the digital world. In order to ensure the resource utilization of edge computing, it is usually necessary to allocate resources among mobile devices. Many scholars have applied auction theory to resource allocation in mobile edge networks and achieved specific results in recent years. However, most existing studies unilaterally consider the utility of users or edge servers, and few jointly consider social welfare and user selection priorities, which cannot maximize the overall utility of mobile edge networks. This paper proposes a multi-round auction algorithm that pursues maximizing social welfare in a multi-user–multi-server network. The algorithm combines combinatorial auction with double auction, allowing users and servers to adjust their bids and asking prices in each auction round. To enhance the generalizability of the study, the algorithm takes full advantage of the flexibility of the auction mechanism and considers the users’ demand for different servers, the variability of bidding strategies, and the variability of the total resources of edge servers so that the users can obtain resources with higher priority and improve social welfare. To ensure that social welfare is maximized, we adopt the Vickrey–Clarke–Groves (VCG) payment mechanism and consider the effect of network externalities. Extensive comparative experiments show that the algorithm can effectively improve social welfare and maximize the overall benefits of the mobile edge network.



中文翻译:

基于多轮拍卖的边缘计算资源分配:最大化社会福利

5G NR-Lite 技术对高移动性条件下的低延迟、高速和海量连接场景的响应仍然不足。基于 4G 和 5G,6G 网络将致力于更全面地支持物联网 (IoT) 范式,致力于为数字世界提供安全的无线计算。为了保证边缘计算的资源利用率,通常需要在移动设备之间分配资源。近年来,许多学者将拍卖理论应用于移动边缘网络的资源分配,并取得了具体成果。然而,现有研究大多单方面考虑用户或边缘服务器的效用,很少联合考虑社会福利和用户选择优先级,无法最大化移动边缘网络的整体效用。本文提出了一种在多用户-多服务器网络中追求社会福利最大化的多轮拍卖算法。该算法结合了组合拍卖和双重拍卖,允许用户和服务器在每一轮拍卖中调整他们的出价和要价。为了增强研究的普适性,该算法充分利用了拍卖机制的灵活性,并考虑了用户对不同服务器的需求、竞价策略的可变性以及边缘服务器总资源的可变性,从而使用户可以获得更高优先级的资源,提高社会福利。为确保社会福利最大化,我们采用 Vickrey–Clarke–Groves (VCG) 支付机制并考虑网络外部性的影响。

更新日期:2022-11-04
down
wechat
bug