ADAFT:SDN大规模流表的适应性深度聚合存储架构
ADAFT:an storage architecture of large-scale SDN flow tables based on adaptive deep aggregations
熊兵 1袁月 1赵锦元 2赵宝康 3何施茗 1张锦1
作者信息
- 1. 长沙理工大学计算机与通信工程学院,湖南 长沙 410114
- 2. 长沙师范学院信息科学与工程学院,湖南 长沙 410199
- 3. 国防科技大学计算机学院,湖南 长沙 410073
- 折叠
摘要
为解决软件定义网络(SDN)数据平面中的三态内容可寻址存储器(TCAM)资源紧张问题,提出了一种基于内容表项树的SDN流表深度聚合方法,进而构建一种SDN大规模流表的适应性深度聚合存储架构ADAFT.该架构放宽了聚合表项之间的汉明距离要求,构建内容表项树聚合动作集不同的流表项,显著提高了流表聚合程度.设计了一种TCAM装载率感知的内容表项树动态限高机制,以降低流表查找开销.同时,提出了一种TCAM装载率感知的表项聚合适应性选择策略,以均衡流表聚合程度和查找开销.实验结果表明,ADAFT架构的流表压缩率明显高于现有方法,最高可达65.74%.
Abstract
To solve the problem of resource shortage of ternary content addressable memory(TCAM)in the data plane of software defined network(SDN),a deep flow table aggregation method was proposed based on content entry trees,and a storage architecture of large-scale SDN flow tables named ADAFT was established.The architecture relaxed the Ham-ming distance requirement between ag-gregated flow entries,and a content entry tree was constructed to aggregate flow entries with different action sets,for significantly en-hancing the aggregation degree of flow tables.Then a dynamic limi-tation mechanism was designed for the height of content entry trees based on the awareness of TCAM load ratio,to mini-mize the lookup overhead of aggregated flow tables.Meanwhile,an adaptive selec-tion strategy of flow entry aggrega-tion was presented in the light of TCAM load ratio,to strike a balance between the aggregation degree and lookup over-head of flow tables.Experimental results indicate that the ADAFT architecture achieves much higher flow table com-pression ratios up to 65.74% than existing methods.
关键词
软件定义网络/SDN大规模流表/内容表项树/适应性深度聚合/TCAM装载率感知Key words
software defined network/large-scale SDN flow table/content entry tree/adaptive deep aggregation/TCAM load ratio awareness引用本文复制引用
基金项目
国家自然科学基金(U22B2005)
国家自然科学基金(61972412)
国家自然科学基金(62272062)
国家重点研发计划(2022YFB2901204)
湖南省自然科学基金(2023JJ30053)
湖南省自然科学基金(2021JJ30456)
湖南省教育厅项目(22A0232)
湖南省教育厅项目(23A0735)
湖南省教育厅项目(22B0300)
湖南省研究生科研创新项目(CX20230913)
出版年
2024
国家科技期刊平台
NSTL
万方数据