基于多维资源感知的算力网络SFC映射策略  

SFC Mapping Strategy for Computing Power Networks Based on Multi-Dimensional Resource Perception

作　　者：吴文鹏 刘晓静 林克全 黄翔 巫泽东 WU Wenpeng;LIU Xiaojing;LIN Kequan;HUANG Xiang;WU Zedong(China Southern Power Grid Co.,Ltd.,Guangzhou,Guangdong 510623,China)

机构地区：[1]中国南方电网有限责任公司,广东广州510623

出　　处：《自动化应用》2025年第6期246-248,共3页Automation Application

基　　金：中国南方电网有限责任公司科技项目(ZBKJXM20232454)。

摘　　要：面对算力网络中日益增长的数据流量和多样化的服务需求,高效、灵活的资源协调成为关键挑战,但现有静态映射算法在处理动态变化的网络环境时往往表现不足。为解决该问题,提出了一种基于多维资源感知的SFC映射策略(SFC-MA)。该策略通过分布式协调,实时监测网络资源状况,准确预测服务需求,并动态优化SFC映射。实验结果表明,与传统的静态映射算法相比,SFC-MA在资源利用率、系统吞吐量和资源碎片化方面均取得了显著提升,能更好地适应动态变化的网络环境,从而提高服务质量并降低运营成本。Faced with the increasing data traffic and diverse service demands in computing power networks,efficient and flexible resource coordination has become a key challenge.However,existing static mapping algorithms often perform poorly in dealing with dynamically changing network environments.To address this issue,this paper proposes a multi-dimensional resource aware SFC-Mapping Strategy(SFC-MA).This strategy utilizes distributed coordination to monitor network resource status in real-time,accurately predict service demands,and dynamically optimize SFC mapping.The experimental results show that compared with traditional static mapping algorithms,SFC-MA has achieved significant improvements in resource utilization,system throughput,and resource fragmentation,and can better adapt to dynamically changing network environments,thereby improving service quality and reducing operating costs.

关 键 词：多维资源感知 监测网络 SFC映射机制 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]