当前位置: X-MOL 学术IEEE Open J. Commun. Soc. › 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.)
Mobility-Aware Energy Optimization in Hosts Selection for Computation Offloading in Multi-Access Edge Computing
IEEE Open Journal of the Communications Society Pub Date : 2020-07-15 , DOI: 10.1109/ojcoms.2020.3008485
Shanmuganathan Thananjeyan , Chien Aun Chan , Elaine Wong , Ampalavanapillai Nirmalathas

Multi-access edge computing (MEC) has been proposed as an approach capable of addressing latency and bandwidth issues in application computation offloading to extend the capabilities beyond the computational and storage limitations of mobile devices. However, there is a critical challenge in MEC to maintain the service continuity between the offloaded user application that is running on the MEC host and the mobile device when a user is moving from radio node to radio node. Furthermore, energy consumption of application computation offloading is an important consideration for MEC service providers in terms of operational costs. Therefore, we formulate the MEC host selection and user application migration problem as a shortest path problem of network energy minimization. We simulate the problem in a hierarchical MEC network deployment environment. We also propose the metric, computational intensity (CI), that can be used by MEC service providers to address the MEC host selection problem. Our results show that with the increment of CI, the selection of MEC hosts tends to move toward level 3 (central deployment) due to energy efficiency and then return to the deployment at level 1 (radio node level) due to latency constraint of the user application. We show that with high accuracy in predicting the user mobility and the available resources in the MEC network, latency- and mobility-aware MEC host selection and user application migration can be pre-calculated to improve response time and energy efficiency.

中文翻译:

多访问边缘计算中用于计算卸载的主机选择中的移动感知能源优化

已经提出了多访问边缘计算(MEC)作为一种能够解决应用计算卸载中的等待时间和带宽问题以将能力扩展到移动设备的计算和存储限制之外的方法。但是,在MEC中,当用户从无线节点移动到无线节点时,要维护在MEC主机上运行的已卸载用户应用程序与移动设备之间的服务连续性是一项严峻的挑战。此外,就运营成本而言,应用计算卸载的能耗是MEC服务提供商的重要考虑因素。因此,我们将MEC主机选择和用户应用程序迁移问题公式化为网络能量最小化的最短路径问题。我们在分层MEC网络部署环境中模拟该问题。我们还提出了度量,计算强度(CI),MEC服务提供商可以使用它来解决MEC主机选择问题。我们的结果表明,随着CI的增加,由于能效的原因,MEC主机的选择趋于向第3级(中央部署)移动,然后由于用户的延迟约束而返回到第1级(无线电节点级别)的部署应用。我们显示,在预测用户移动性和MEC网络中可用资源的准确性很高的情况下,可以预先计算延迟和移动性感知的MEC主机选择以及用户应用程序迁移,以改善响应时间和能源效率。MEC主机的选择由于能效而趋向于第3级(中央部署),然后由于用户应用程序的延迟约束而返回到第1级(无线电节点级别)的部署。我们显示,在预测用户移动性和MEC网络中可用资源的准确性很高的情况下,可以预先计算延迟和移动性MEC主机选择以及用户应用程序迁移,以改善响应时间和能源效率。MEC主机的选择由于能效而趋向于第3级(中央部署),然后由于用户应用程序的延迟约束而返回到第1级(无线电节点级别)的部署。我们显示,在预测用户移动性和MEC网络中可用资源的准确性很高的情况下,可以预先计算延迟和移动性感知的MEC主机选择以及用户应用程序迁移,以改善响应时间和能源效率。
更新日期:2020-08-11
down
wechat
bug