{"title":"SD-FAST: SDN中的分组重路由体系结构","authors":"M. A. Moyeen, Fangye Tang, Dipon Saha, I. Haque","doi":"10.23919/CNSM46954.2019.9012703","DOIUrl":null,"url":null,"abstract":"Communication link failure is common in any network. In Software Defined Network (SDN), protection-based recovery scheme reduces the failure recovery delay by installing alternative routes at the data plane switches. We can deploy Fast Failure Group (FFG) of OpenFlow protocol if a switch has an alternative path towards the destination; otherwise, the switch can use crankback approach to send the affected traffic towards the traversed route to find an alternative path. These existing recovery schemes force every packet to traverse a chain of matching tables even in the absence of a link failure, which impacts packet processing time and end-to-end delay. In this paper, we propose a packet rerouting architecture, called SD-FAST, that invokes recovery scheme only after facing failure and reduces both the packet processing and crankback backtracking time. We evaluate SD-FAST in Mininet, considering real and simulated traffic on real network topologies. The evaluation results confirm that SD-FAST can reduce around 73% crankback backtracking time and 64% delay compared to its counterparts.","PeriodicalId":273818,"journal":{"name":"2019 15th International Conference on Network and Service Management (CNSM)","volume":"20 17-18 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2019-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"5","resultStr":"{\"title\":\"SD-FAST: A Packet Rerouting Architecture in SDN\",\"authors\":\"M. A. Moyeen, Fangye Tang, Dipon Saha, I. Haque\",\"doi\":\"10.23919/CNSM46954.2019.9012703\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Communication link failure is common in any network. In Software Defined Network (SDN), protection-based recovery scheme reduces the failure recovery delay by installing alternative routes at the data plane switches. We can deploy Fast Failure Group (FFG) of OpenFlow protocol if a switch has an alternative path towards the destination; otherwise, the switch can use crankback approach to send the affected traffic towards the traversed route to find an alternative path. These existing recovery schemes force every packet to traverse a chain of matching tables even in the absence of a link failure, which impacts packet processing time and end-to-end delay. In this paper, we propose a packet rerouting architecture, called SD-FAST, that invokes recovery scheme only after facing failure and reduces both the packet processing and crankback backtracking time. We evaluate SD-FAST in Mininet, considering real and simulated traffic on real network topologies. The evaluation results confirm that SD-FAST can reduce around 73% crankback backtracking time and 64% delay compared to its counterparts.\",\"PeriodicalId\":273818,\"journal\":{\"name\":\"2019 15th International Conference on Network and Service Management (CNSM)\",\"volume\":\"20 17-18 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2019-10-01\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"5\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"2019 15th International Conference on Network and Service Management (CNSM)\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.23919/CNSM46954.2019.9012703\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"2019 15th International Conference on Network and Service Management (CNSM)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.23919/CNSM46954.2019.9012703","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 5
摘要
通信链路故障在任何网络中都是常见的。在SDN (Software Defined Network)中,基于保护的恢复方案通过在数据平面交换机上安装备选路由来减少故障恢复的延迟。如果交换机到目的地有替代路径,我们可以部署OpenFlow协议的快速故障组(FFG);否则,交换机可以使用曲柄倒转的方式将受影响的流量发送到经过的路由上,以寻找替代路径。这些现有的恢复方案强制每个数据包遍历匹配表链,即使在没有链路故障的情况下,这也会影响数据包处理时间和端到端延迟。在本文中,我们提出了一种数据包重路由架构,称为SD-FAST,它只在遇到故障时调用恢复方案,从而减少了数据包处理和反向回溯时间。我们在Mininet中评估了SD-FAST,考虑了真实网络拓扑上的真实和模拟流量。评估结果证实,与同类产品相比,SD-FAST可以减少约73%的曲轴回退时间和64%的延迟。
Communication link failure is common in any network. In Software Defined Network (SDN), protection-based recovery scheme reduces the failure recovery delay by installing alternative routes at the data plane switches. We can deploy Fast Failure Group (FFG) of OpenFlow protocol if a switch has an alternative path towards the destination; otherwise, the switch can use crankback approach to send the affected traffic towards the traversed route to find an alternative path. These existing recovery schemes force every packet to traverse a chain of matching tables even in the absence of a link failure, which impacts packet processing time and end-to-end delay. In this paper, we propose a packet rerouting architecture, called SD-FAST, that invokes recovery scheme only after facing failure and reduces both the packet processing and crankback backtracking time. We evaluate SD-FAST in Mininet, considering real and simulated traffic on real network topologies. The evaluation results confirm that SD-FAST can reduce around 73% crankback backtracking time and 64% delay compared to its counterparts.