【摘要】：隨著社會(huì)經(jīng)濟(jì)的不斷發(fā)展,軟件定義聯(lián)網(wǎng)作為一種新型網(wǎng)絡(luò)體系架構(gòu)一直備受人們關(guān)注。SDN具有高度的網(wǎng)絡(luò)抽象、新穎的集中智能和良好的可編程性特點(diǎn),然而SDN網(wǎng)絡(luò)仍然避免不了傳統(tǒng)網(wǎng)絡(luò)出現(xiàn)的網(wǎng)絡(luò)故障現(xiàn)象。作為全網(wǎng)的控制和管理中心的SDN控制器的癱瘓或故障,容易造成全網(wǎng)服務(wù)的中斷等不利影響。為此,研究SDN網(wǎng)絡(luò)故障監(jiān)控和故障恢復(fù)技術(shù)具有較大的實(shí)際應(yīng)用價(jià)值。本文分析了SDN網(wǎng)絡(luò)體系結(jié)構(gòu),研究了控制器和SDN交換機(jī)的工作流程,梳理了威脅SDN網(wǎng)絡(luò)的安全性因素,總結(jié)了SDN網(wǎng)絡(luò)故障范圍,確定了故障類型。結(jié)合了OpenFlow消息類型,從SDN網(wǎng)絡(luò)數(shù)據(jù)平面和控制平面兩個(gè)角度,針對(duì)不同的故障類型,歸納出故障監(jiān)控方法。研究了數(shù)據(jù)平面主動(dòng)方式和被動(dòng)方式兩種故障恢復(fù)方法,分析了兩種方法的優(yōu)缺點(diǎn),提出一種主被動(dòng)方式結(jié)合的、面向多個(gè)邏輯SDN控制器的故障恢復(fù)方法,分析了現(xiàn)有比較成熟的多控制器體系和OpenFlow協(xié)議有關(guān)多控制器的搭建要求,基于主備控制器模型,實(shí)現(xiàn)了面向多個(gè)邏輯SDN控制器故障恢復(fù)方法。本文使用Big Switch Networks公司核心的企業(yè)級(jí)控制器Floodlight和OpenvSwitch軟件交換機(jī),采用物理實(shí)驗(yàn)設(shè)備和仿真軟件Mininet模擬,結(jié)合Wireshark和Iperf等測(cè)試工具,對(duì)SDN網(wǎng)絡(luò)故障監(jiān)控和恢復(fù)進(jìn)行測(cè)試與驗(yàn)證。測(cè)試結(jié)果表明,相關(guān)的功能模塊可以實(shí)現(xiàn)對(duì)這些故障的監(jiān)控和恢復(fù)。
[Abstract]:With the continuous development of social economy, software definition networking, as a new network architecture, has been paid more and more attention. SDN has the characteristics of high network abstraction, novel centralized intelligence and good programming. However, SDN network can not avoid the network failure of traditional network. The paralysis or failure of SDN controller, which is the control and management center of the whole network, is easy to cause the interruption of the whole network service and so on. Therefore, it is of great practical application value to study SDN network fault monitoring and fault recovery technology. In this paper, the architecture of SDN network is analyzed, the workflow of controller and SDN switch is studied, the security factors that threaten SDN network are combed, the fault range of SDN network is summarized, and the fault type is determined. Combined with OpenFlow message type, the fault monitoring method is summarized from two angles of SDN network data plane and control plane, according to different fault types. Two fault recovery methods, active mode and passive mode, are studied, the advantages and disadvantages of the two methods are analyzed, and a fault recovery method for multiple logical SDN controllers based on active and passive mode is proposed. The requirements of multi-controller construction for the existing mature multi-controller system and OpenFlow protocol are analyzed. Based on the main and standby controller model, the fault recovery method for multiple logic SDN controllers is implemented. In this paper, the core enterprise controller Floodlight and OpenvSwitch software switch of Big Switch Networks Company, physical experiment equipment and simulation software Mininet simulation, combined with Wireshark and Iperf and other test tools, are used to test and verify the fault monitoring and recovery of SDN network. The test results show that the related functional modules can monitor and recover these faults.
【學(xué)位授予單位】：南京郵電大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2017
【分類號(hào)】：TP393.0

【參考文獻(xiàn)】

相關(guān)期刊論文 前5條

1 侯樂(lè);汪碩;林毅凱;姚博韜;劉江;;基于SDN的鏈路故障恢復(fù)[J];電信科學(xué);2015年06期

2 沈蘇彬;;軟件定義聯(lián)網(wǎng)的建模與分析[J];南京郵電大學(xué)學(xué)報(bào)(自然科學(xué)版);2014年03期

3 林萍萍;畢軍;胡虹雨;蔣小可;;一種面向SDN域內(nèi)控制平面可擴(kuò)展性的機(jī)制[J];小型微型計(jì)算機(jī)系統(tǒng);2013年09期

4 左青云;陳鳴;趙廣松;邢長(zhǎng)友;張國(guó)敏;蔣培成;;基于OpenFlow的SDN技術(shù)研究[J];軟件學(xué)報(bào);2013年05期

5 張民貴;劉斌;;IP網(wǎng)絡(luò)的快速故障恢復(fù)[J];電子學(xué)報(bào);2008年08期

相關(guān)會(huì)議論文 前1條

1 吳舜;蘇丹;吳佳;李坤;杜劍雯;聶正璞;;OpenFlow分布式多控制器設(shè)計(jì)與實(shí)現(xiàn)[A];2013電力行業(yè)信息化年會(huì)論文集[C];2013年

相關(guān)碩士學(xué)位論文 前5條

1 李凡卡;基于SDN控制器集群的集中管控技術(shù)[D];電子科技大學(xué);2016年

2 姚龍;軟件定義網(wǎng)絡(luò)控制器容量及部署問(wèn)題研究[D];中國(guó)科學(xué)技術(shù)大學(xué);2015年

3 馬文婷;基于OpenFlow的SDN控制器關(guān)鍵技術(shù)研究[D];北京郵電大學(xué);2015年

4 馬飛翔;基于SDN技術(shù)的網(wǎng)絡(luò)故障管理系統(tǒng)設(shè)計(jì)與實(shí)現(xiàn)[D];北京交通大學(xué);2014年

5 張吉興;基于SDN的控制器集群技術(shù)的研究[D];電子科技大學(xué);2014年

，

本文編號(hào)：2496712