恶意流量检测与边缘环境容器迁移

Malicious Traffic Detection and Container Migration Method for Edge Environments

介绍一种基于恶意流量检测的容器服务预迁移方法,旨在增强边缘计算环境的数据安全性、服务可用性和网络可靠性。将边缘网络抽象为图结构,以描述设备和网络流的特性,包括计算资源、存储资源、物理位置和网络流属性。引入了一个恶意流量检测模型(MTDG),利用图采样与聚合算法实时对网络流进行分类。将容器服务迁移视为多目标优化问题,综合考虑了能耗和网络负载均衡,提出了一种基于NSGA-Ⅱ的容器服务迁移策略,通过容器实时迁移技术,在不中断服务的情况下隔离和维护异常节点。仿真实验结果表明,该策略降低了能耗和负载均衡系数,验证了其可行性和有效性。

This article introduces a container service pre-migration method based on malicious traffic detection,aiming to enhance data security,service availability,and network reliability in edge computing environments.The article abstracts the edge network into a graph structure to describe the characteristics of devices and network flows in detail,including computing resources,storage resources,physical locations,and network flow attributes.A malicious traffic detection model(MTDG)is introduced to classify network flows in real time using graph sampling and aggregation algorithms.The article regards container service migration as a multi-objective optimization problem,comprehensively considers energy consumption and network load balancing,and proposes a container service migration strategy based on NSGA-II,which uses container live migration technology to isolate without interrupting services.and maintain abnormal nodes.Simulation experiment results show that this strategy reduces energy consumption and load balancing coefficients,verifying its feasibility and effectiveness.