Qinglei Qi, Wendong Wang, Xiangyang Gong, Xirong Que
{"title":"组合SDN转发单元的分析模型","authors":"Qinglei Qi, Wendong Wang, Xiangyang Gong, Xirong Que","doi":"10.23919/INM.2017.7987332","DOIUrl":null,"url":null,"abstract":"Recent studies have shown that the flow table size of hardware SDN switch cannot match the number of concurrent flows. Combined SDN Forwarding Element (CFE), which comprises software switch and hardware switch, becomes an alternative approach for tackling this problem. Because software switch has lower lookup speed than hardware switch, different proportions of traffic allocated to software switches in CFE have different effects on the delay bounds of all flows entering CFE. As delay-guarantee is a nontrivial task for network providers, especially with the increasing number of delay-sensitive applications, a model to analyze the delay bound given a flow allocation in CFE is important. With the one-to-one correspondence between flow allocation and rules placement solution, the analytical model can be used to evaluate and compare rules placement solutions and provide a basis for designing better rules placement solution in CFE. In this paper, we propose an analytical model for CFE based on network calculus, and then validate this model through simulations in NS-3.","PeriodicalId":119633,"journal":{"name":"2017 IFIP/IEEE Symposium on Integrated Network and Service Management (IM)","volume":"64 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2017-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"An analytical model for combined SDN Forwarding Element\",\"authors\":\"Qinglei Qi, Wendong Wang, Xiangyang Gong, Xirong Que\",\"doi\":\"10.23919/INM.2017.7987332\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Recent studies have shown that the flow table size of hardware SDN switch cannot match the number of concurrent flows. Combined SDN Forwarding Element (CFE), which comprises software switch and hardware switch, becomes an alternative approach for tackling this problem. Because software switch has lower lookup speed than hardware switch, different proportions of traffic allocated to software switches in CFE have different effects on the delay bounds of all flows entering CFE. As delay-guarantee is a nontrivial task for network providers, especially with the increasing number of delay-sensitive applications, a model to analyze the delay bound given a flow allocation in CFE is important. With the one-to-one correspondence between flow allocation and rules placement solution, the analytical model can be used to evaluate and compare rules placement solutions and provide a basis for designing better rules placement solution in CFE. In this paper, we propose an analytical model for CFE based on network calculus, and then validate this model through simulations in NS-3.\",\"PeriodicalId\":119633,\"journal\":{\"name\":\"2017 IFIP/IEEE Symposium on Integrated Network and Service Management (IM)\",\"volume\":\"64 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2017-05-01\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"2017 IFIP/IEEE Symposium on Integrated Network and Service Management (IM)\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.23919/INM.2017.7987332\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"2017 IFIP/IEEE Symposium on Integrated Network and Service Management (IM)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.23919/INM.2017.7987332","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0
摘要
最近的研究表明,硬件SDN交换机的流表大小不能匹配并发流的数量。组合式SDN转发单元(Combined SDN Forwarding Element, CFE)由软件交换机和硬件交换机组成,成为解决这一问题的另一种方法。由于软件交换机的查找速度比硬件交换机慢,所以在CFE中分配给软件交换机的流量的不同比例对所有进入CFE的流量的延迟界有不同的影响。对于网络提供商来说,延迟保证是一项非常重要的任务,特别是随着对延迟敏感的应用数量的增加,在CFE中分析给定流量分配的延迟边界模型是非常重要的。由于流分配与规则放置方案之间存在一一对应关系,该分析模型可用于评估和比较规则放置方案,为CFE中设计更好的规则放置方案提供依据。本文提出了一种基于网络演算的CFE分析模型,并在NS-3中进行了仿真验证。
An analytical model for combined SDN Forwarding Element
Recent studies have shown that the flow table size of hardware SDN switch cannot match the number of concurrent flows. Combined SDN Forwarding Element (CFE), which comprises software switch and hardware switch, becomes an alternative approach for tackling this problem. Because software switch has lower lookup speed than hardware switch, different proportions of traffic allocated to software switches in CFE have different effects on the delay bounds of all flows entering CFE. As delay-guarantee is a nontrivial task for network providers, especially with the increasing number of delay-sensitive applications, a model to analyze the delay bound given a flow allocation in CFE is important. With the one-to-one correspondence between flow allocation and rules placement solution, the analytical model can be used to evaluate and compare rules placement solutions and provide a basis for designing better rules placement solution in CFE. In this paper, we propose an analytical model for CFE based on network calculus, and then validate this model through simulations in NS-3.