A Study of Improving Traffic Distribution Mechanism in Software-Defined Networking

Ukrist Srisamarn, L. Pradittasnee
{"title":"A Study of Improving Traffic Distribution Mechanism in Software-Defined Networking","authors":"Ukrist Srisamarn, L. Pradittasnee","doi":"10.1109/TENCON.2018.8650302","DOIUrl":null,"url":null,"abstract":"Recently, computer networks become one of the main components of the business processes in most organizations in the world. The size and complexity of network systems also continue to rapidly grow every year. When the networks become larger, many traffic flows are likely to propagate in the network at the same time. This event can lead to multiple performance problems, such as congestion, and slow response time. Traffic distribution mechanism is one of the main solutions for solving congestion problem. It aims to reduce the probability of congestion by splitting the traffic load into multiple paths. Furthermore, applying traffic distribution technique with Software-Defined Networking (SDN) is a promising solution that can effectively support both large-scale network implementation and also does not add high level of complexity to the network system. In this paper, the traffic distribution mechanism based on modified OpenFlow is proposed. The proposed method aims to decrease response time that a SDN switch must wait for the control message back from the SDN controller. It can be achieved by allowing the SDN controller to add the new flow entries for distributing the traffic load before the packet arrives at the SDN switch. The simulation experiments are conducted in Mininet to compare the performance between the proposed traffic distribution mechanism (TDm) and the traffic distribution mechanism based on OpenFlow (TDo). The results from the experiments show that TDm can provide much better performance in term of End-to-End delay (E2E delay) and Packet delivery ratio (PDR) than TDo in most of the experiment scenarios. However, both traffic distribution mechanisms (TDo and TDm) cannot provide good performance in term of E2E delay and PDR when the data transmission rate equals or higher than 500 Mbps. As a result, Time-Sleep mechanism is proposed to implement in SDN controller. The results from the simulation experiments show that Time-Sleep mechanism can notably improve the overall performance of both TDo and TDm even when the data transmission rate is high.","PeriodicalId":132900,"journal":{"name":"TENCON 2018 - 2018 IEEE Region 10 Conference","volume":"169 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2018-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"1","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"TENCON 2018 - 2018 IEEE Region 10 Conference","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/TENCON.2018.8650302","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 1

Abstract

Recently, computer networks become one of the main components of the business processes in most organizations in the world. The size and complexity of network systems also continue to rapidly grow every year. When the networks become larger, many traffic flows are likely to propagate in the network at the same time. This event can lead to multiple performance problems, such as congestion, and slow response time. Traffic distribution mechanism is one of the main solutions for solving congestion problem. It aims to reduce the probability of congestion by splitting the traffic load into multiple paths. Furthermore, applying traffic distribution technique with Software-Defined Networking (SDN) is a promising solution that can effectively support both large-scale network implementation and also does not add high level of complexity to the network system. In this paper, the traffic distribution mechanism based on modified OpenFlow is proposed. The proposed method aims to decrease response time that a SDN switch must wait for the control message back from the SDN controller. It can be achieved by allowing the SDN controller to add the new flow entries for distributing the traffic load before the packet arrives at the SDN switch. The simulation experiments are conducted in Mininet to compare the performance between the proposed traffic distribution mechanism (TDm) and the traffic distribution mechanism based on OpenFlow (TDo). The results from the experiments show that TDm can provide much better performance in term of End-to-End delay (E2E delay) and Packet delivery ratio (PDR) than TDo in most of the experiment scenarios. However, both traffic distribution mechanisms (TDo and TDm) cannot provide good performance in term of E2E delay and PDR when the data transmission rate equals or higher than 500 Mbps. As a result, Time-Sleep mechanism is proposed to implement in SDN controller. The results from the simulation experiments show that Time-Sleep mechanism can notably improve the overall performance of both TDo and TDm even when the data transmission rate is high.
软件定义网络中改进流量分配机制的研究
近年来,计算机网络已成为世界上大多数组织的业务流程的主要组成部分之一。网络系统的规模和复杂性也在逐年快速增长。当网络规模变大时,许多流量可能同时在网络中传播。此事件可能导致多种性能问题,例如拥塞和缓慢的响应时间。交通分配机制是解决交通拥堵问题的主要方法之一。它的目的是通过将流量负载分成多条路径来降低拥塞的概率。此外,将流量分配技术应用于软件定义网络(SDN)是一种很有前途的解决方案,它既可以有效地支持大规模网络实施,又不会增加网络系统的高度复杂性。本文提出了基于改进OpenFlow的流量分配机制。提出的方法旨在减少SDN交换机必须等待从SDN控制器返回的控制消息的响应时间。这可以通过允许SDN控制器在数据包到达SDN交换机之前添加新的流项来分配流量负载来实现。在Mininet中进行了仿真实验,比较了所提出的流量分配机制(TDm)和基于OpenFlow的流量分配机制(TDo)的性能。实验结果表明,在大多数实验场景下,TDm在端到端延迟(E2E延迟)和分组分发率(PDR)方面都比TDo具有更好的性能。但是,当数据传输速率大于等于500mbps时,TDo和TDm两种流量分配机制在端到端延迟和PDR方面都不能提供良好的性能。因此,提出了在SDN控制器中实现时间-睡眠机制。仿真实验结果表明,即使在数据传输速率较高的情况下,Time-Sleep机制也能显著提高TDo和TDm的整体性能。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
求助全文
约1分钟内获得全文 求助全文
来源期刊
自引率
0.00%
发文量
0
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
copy
已复制链接
快去分享给好友吧!
我知道了
右上角分享
点击右上角分享
0
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术官方微信