本文選題：內(nèi)容分發(fā)網(wǎng)絡(luò)　切入點(diǎn)：全局服務(wù)器負(fù)載均衡　出處：《華南理工大學(xué)》2014年碩士論文　論文類(lèi)型：學(xué)位論文

【摘要】：21世紀(jì)互聯(lián)網(wǎng)迅速發(fā)展普及，人們通過(guò)上網(wǎng)獲取信息，與朋友通信交流，觀看電影等�；ヂ�(lián)網(wǎng)極大地提高了人們的工作效率，豐富了人們的精神生活。與此同時(shí)，，隨著使用互聯(lián)網(wǎng)的人越來(lái)越多，互聯(lián)網(wǎng)的內(nèi)容也以幾何增長(zhǎng)方式迅速膨脹，互聯(lián)網(wǎng)變得越來(lái)越擁擠。由于缺乏必要的流量管理和疏通手段，骨干帶寬被迅速消耗掉，IP網(wǎng)絡(luò)流量秩序?yàn)l于失衡。基于這個(gè)問(wèn)題，互聯(lián)網(wǎng)提出一種叫做CDN的解決方案，CDN的全稱(chēng)是ContentDeliveryNetwork，即內(nèi)容分發(fā)網(wǎng)絡(luò)。其目的是通過(guò)在現(xiàn)有的Internet中增加一層新的網(wǎng)絡(luò)架構(gòu)，將網(wǎng)站的內(nèi)容發(fā)布到最接近用戶(hù)的網(wǎng)絡(luò)"邊緣"，使用戶(hù)可以就近取得所需的內(nèi)容，解決Internet網(wǎng)絡(luò)擁擠的狀況，提高用戶(hù)訪(fǎng)問(wèn)網(wǎng)站的響應(yīng)速度。從技術(shù)上全面解決由于網(wǎng)絡(luò)帶寬小、用戶(hù)訪(fǎng)問(wèn)量大、網(wǎng)點(diǎn)分布不均等原因所造成的用戶(hù)訪(fǎng)問(wèn)網(wǎng)站響應(yīng)速度慢的問(wèn)題。隨著CDN技術(shù)的不斷發(fā)展，現(xiàn)在的CDN解決方案已經(jīng)不在滿(mǎn)足于簡(jiǎn)單的通過(guò)增加邊緣服務(wù)來(lái)提升用戶(hù)體驗(yàn)，而是需要一套智能的，實(shí)時(shí)感應(yīng)用戶(hù)服務(wù)質(zhì)量和實(shí)時(shí)網(wǎng)絡(luò)質(zhì)量的系統(tǒng)。通過(guò)獲取到實(shí)時(shí)的網(wǎng)絡(luò)情況，分析用戶(hù)訪(fǎng)問(wèn)的最佳鏈路，為用戶(hù)提供最好的節(jié)點(diǎn)服務(wù)，從而提升用戶(hù)的訪(fǎng)問(wèn)質(zhì)量。 本文選擇“內(nèi)容分發(fā)網(wǎng)絡(luò)調(diào)度系統(tǒng)的研究與實(shí)現(xiàn)”作為研究課題，通過(guò)研究?jī)?nèi)容分發(fā)系統(tǒng)的調(diào)度策略，通過(guò)基于IP歸屬地區(qū)和運(yùn)營(yíng)商的分析和內(nèi)容分發(fā)系統(tǒng)各個(gè)節(jié)點(diǎn)的實(shí)時(shí)帶寬IO統(tǒng)計(jì)，綜合分析從而實(shí)現(xiàn)內(nèi)容分發(fā)網(wǎng)絡(luò)智能調(diào)度。本系統(tǒng)研究和實(shí)現(xiàn)的關(guān)鍵點(diǎn)在于，如何能夠智能地進(jìn)行網(wǎng)絡(luò)流量調(diào)度和節(jié)點(diǎn)間的負(fù)載均衡調(diào)度。最后設(shè)計(jì)實(shí)現(xiàn)CDN全局負(fù)載均衡調(diào)度系統(tǒng)，實(shí)現(xiàn)網(wǎng)站節(jié)點(diǎn)間的智能調(diào)度，提升用戶(hù)的訪(fǎng)問(wèn)資源下載速度，提高用戶(hù)的訪(fǎng)問(wèn)體驗(yàn)。另外也保障各服務(wù)節(jié)點(diǎn)的負(fù)載均衡，從而使服務(wù)資源能夠充分利用，提高效率。
[Abstract]:In 21th century, the Internet was rapidly developed and popularized. People get information through the Internet, communicate with friends, watch movies, etc. The Internet has greatly improved people's working efficiency and enriched people's spiritual life. At the same time, As more and more people use the Internet, the content of the Internet expands rapidly in a geometric way, and the Internet becomes more and more crowded. Due to the lack of necessary traffic management and dredging means, Backbone bandwidth is rapidly consumed and IP network traffic order is on the verge of imbalance. The Internet proposes a solution called CDN, the full name of which is content delivery Network, which aims to add a new layer of network architecture to the existing Internet. By publishing the content of the website to the "edge" of the network closest to the user, the user can obtain the required content in the vicinity, solve the congested condition of the Internet network, and improve the response speed of the user to visit the website. The problem of slow response of users to visit websites caused by the large number of users and uneven distribution of sites. With the development of CDN technology, Today's CDN solution is no longer satisfied with simply enhancing the user experience by adding edge services, but requires an intelligent set of solutions. The system of real-time sensing user quality of service and real-time network quality. By obtaining the real-time network situation, analyzing the best link accessed by the user, providing the best node service for the user, thus improving the user's access quality. In this paper, "the Research and implementation of content Distribution Network scheduling system" is selected as the research topic, and the scheduling strategy of the content Distribution system is studied. Through the analysis based on IP ownership area and operator and the real-time bandwidth IO statistics of each node in the content distribution system, the intelligent scheduling of content distribution network is realized by comprehensive analysis. The key point of this system research and implementation lies in, How to intelligently schedule network traffic and load balancing between nodes. Finally, the CDN global load balancing scheduling system is designed and implemented. In addition, it also ensures the load balance of each service node, so that the service resources can be fully utilized and the efficiency can be improved.
【學(xué)位授予單位】：華南理工大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2014
【分類(lèi)號(hào)】：TP393.02

【參考文獻(xiàn)】

相關(guān)期刊論文 前9條

1 鄭淑暉,侯朝楨;基于自學(xué)習(xí)模式下的網(wǎng)絡(luò)流量控制研究[J];北京理工大學(xué)學(xué)報(bào);2003年02期

2 趙鳳梅;;《電化教育研究》網(wǎng)絡(luò)引文分析研究[J];電化教育研究;2006年10期

3 王超;趙文杰;;IP網(wǎng)絡(luò)帶寬管理技術(shù)及應(yīng)用分析[J];電信技術(shù);2007年05期

4 廖利;陳瑩;;CDN技術(shù)在流媒體中的應(yīng)用[J];福建電腦;2007年08期

5 陳學(xué)鋒,周亞敏,敖青云,白英彩;分布式實(shí)時(shí)網(wǎng)絡(luò)監(jiān)測(cè)系統(tǒng)的設(shè)計(jì)與實(shí)現(xiàn)[J];計(jì)算機(jī)工程;2002年06期

6 畢經(jīng)平,吳起,李忠誠(chéng);Internet中的包延遲分布與包丟失關(guān)系研究[J];計(jì)算機(jī)工程;2002年10期

7 宋迎梅;沈富可;;基于P4P的CDN系統(tǒng)中內(nèi)容分發(fā)節(jié)點(diǎn)選擇算法[J];計(jì)算機(jī)應(yīng)用;2009年11期

8 徐恪;;P2P流量的監(jiān)控與管理[J];中國(guó)教育網(wǎng)絡(luò);2006年07期

9 姚正;;計(jì)算機(jī)發(fā)展趨勢(shì)展望[J];商情(教育經(jīng)濟(jì)研究);2008年01期

相關(guān)博士學(xué)位論文 前2條

1 霍英;基于超節(jié)點(diǎn)體系結(jié)構(gòu)的P2P覆蓋網(wǎng)優(yōu)化技術(shù)研究[D];中南大學(xué);2011年

2 李軍;大規(guī)模數(shù)字內(nèi)容網(wǎng)絡(luò)分發(fā)關(guān)鍵技術(shù)研究[D];北京郵電大學(xué);2009年

本文編號(hào)：1557150