【摘要】：網(wǎng)絡(luò)流量管理是網(wǎng)絡(luò)管理的重要組成部分,網(wǎng)絡(luò)流量是網(wǎng)絡(luò)中傳輸?shù)臄?shù)據(jù)量,是體現(xiàn)網(wǎng)絡(luò)性能狀況和運(yùn)行狀態(tài)的關(guān)鍵數(shù)據(jù)。通過對網(wǎng)絡(luò)流量進(jìn)行分析研究,不僅能夠起到對網(wǎng)絡(luò)流量的監(jiān)測作用,同時(shí)還能掌握一定的網(wǎng)絡(luò)行為規(guī)律,實(shí)施網(wǎng)絡(luò)流量預(yù)測,對于流量異常情況進(jìn)行及時(shí)告警,以幫助網(wǎng)絡(luò)管理人員排查和定位網(wǎng)絡(luò)故障,并為優(yōu)化網(wǎng)絡(luò)拓?fù)浜瓦M(jìn)行網(wǎng)絡(luò)規(guī)劃提供科學(xué)依據(jù)。本文以網(wǎng)絡(luò)流量為核心,融合了網(wǎng)絡(luò)流量相關(guān)技術(shù),包括網(wǎng)絡(luò)流量監(jiān)測技術(shù)、網(wǎng)絡(luò)流量預(yù)測技術(shù)和網(wǎng)絡(luò)流量異常檢測技術(shù),設(shè)計(jì)并實(shí)現(xiàn)了一個(gè)基于B/S模式且具有良好可用性、可視性和可擴(kuò)展性的網(wǎng)絡(luò)流量監(jiān)測系統(tǒng)。系統(tǒng)采用了基于SNMP協(xié)議的網(wǎng)絡(luò)流量監(jiān)測技術(shù)實(shí)現(xiàn)對網(wǎng)絡(luò)流量數(shù)據(jù)的采集,通過多線程采集方式提高了系統(tǒng)的數(shù)據(jù)采集效率;針對所采集網(wǎng)絡(luò)流量的歷史數(shù)據(jù),采用時(shí)間序列分析方法進(jìn)行數(shù)學(xué)建模,建模過程使用了經(jīng)典的Box-Jenkins方法,包括時(shí)間序列的平穩(wěn)性檢驗(yàn)、非平穩(wěn)時(shí)間序列的平穩(wěn)化、模型定階、模型參數(shù)估計(jì)、模型檢驗(yàn)和模型預(yù)測幾個(gè)主要步驟;使用所建立的ARMA模型實(shí)現(xiàn)對網(wǎng)絡(luò)流量的短期在線預(yù)測,并為網(wǎng)絡(luò)流量自適應(yīng)閾值異常檢測方法中閾值邊界的確定提供了來源。系統(tǒng)能很好地實(shí)時(shí)監(jiān)測目標(biāo)網(wǎng)絡(luò)的網(wǎng)絡(luò)流量,將網(wǎng)絡(luò)流量以清晰的圖表形式進(jìn)行顯示,能準(zhǔn)確預(yù)測未來一段時(shí)期內(nèi)網(wǎng)絡(luò)流量的變化趨勢,并對可能的網(wǎng)絡(luò)流量異常進(jìn)行及時(shí)的檢測和告警。本文所實(shí)現(xiàn)的網(wǎng)絡(luò)流量監(jiān)測系統(tǒng)現(xiàn)已部署到華南理工大學(xué)廣東省計(jì)算機(jī)網(wǎng)絡(luò)重點(diǎn)實(shí)驗(yàn)室的服務(wù)器上,經(jīng)測試系統(tǒng)運(yùn)行良好,達(dá)到了預(yù)期目標(biāo),能滿足校園網(wǎng)針對網(wǎng)絡(luò)流量管理上的實(shí)際需求。
[Abstract]:Network traffic management is an important part of network management. Network traffic is the amount of data transmitted in the network, and it is the key data to reflect the network performance and running state. By analyzing and studying the network traffic, not only can the network traffic be monitored, but also the network behavior law can be grasped, the network traffic forecast can be carried out, and the abnormal traffic situation can be alerted in time. In order to help network managers detect and locate network failures, and provide scientific basis for optimizing network topology and network planning. Based on the core of network traffic, this paper combines network traffic related technologies, including network traffic monitoring technology, network traffic prediction technology and network traffic anomaly detection technology, and designs and implements a B / S model with good usability. Visibility and scalability of network traffic monitoring system. The system adopts the network traffic monitoring technology based on SNMP protocol to collect the network traffic data, improves the efficiency of the data collection by multi-thread acquisition, and aims at the historical data of the collected network traffic. The classical Box-Jenkins method is used in the modeling process, including the stationary test of time series, the stationary of non-stationary time series, the determination of model order, and the estimation of model parameters. There are several main steps of model checking and model prediction, and the established ARMA model is used to realize the short-term on-line prediction of network traffic, which provides a source for the determination of threshold boundary in the adaptive threshold anomaly detection method of network traffic. The system can monitor the network traffic of the target network in real time and display the network traffic in a clear chart form, which can accurately predict the change trend of the network traffic in a period of time in the future. And the possible network traffic anomaly timely detection and alarm. The network traffic monitoring system implemented in this paper has been deployed to the server of Guangdong computer Network key Laboratory of South China University of Technology. It can meet the actual demand of network traffic management in campus network.
【學(xué)位授予單位】：華南理工大學(xué)
【學(xué)位級別】：碩士
【學(xué)位授予年份】：2015
【分類號(hào)】：TP393.06

【參考文獻(xiàn)】

相關(guān)期刊論文 前5條

1 趙軍圣;莊光明;王增桂;;極大似然估計(jì)方法介紹[J];長春理工大學(xué)學(xué)報(bào);2010年06期

2 鄒柏賢,劉強(qiáng);基于ARMA模型的網(wǎng)絡(luò)流量預(yù)測[J];計(jì)算機(jī)研究與發(fā)展;2002年12期

3 孫泳,史忠植;MRTG的研究與部署[J];計(jì)算機(jī)應(yīng)用;2004年03期

4 劉羅曼;;時(shí)間序列平穩(wěn)性檢驗(yàn)[J];沈陽師范大學(xué)學(xué)報(bào)(自然科學(xué)版);2010年03期

5 劉經(jīng)南;曾文憲;徐培亮;;整體最小二乘估計(jì)的研究進(jìn)展[J];武漢大學(xué)學(xué)報(bào)(信息科學(xué)版);2013年05期

相關(guān)博士學(xué)位論文 前1條

1 高波;基于時(shí)間相關(guān)的網(wǎng)絡(luò)流量建模與預(yù)測研究[D];哈爾濱工業(yè)大學(xué);2013年

相關(guān)碩士學(xué)位論文 前1條

1 張久坤;網(wǎng)絡(luò)流量模型與流量預(yù)測技術(shù)研究[D];西安電子科技大學(xué);2012年

，

本文編號(hào)：2210640