本文關(guān)鍵詞： Hadoop 數(shù)據(jù)中心 分布式存儲(chǔ) 分布式計(jì)算 視頻轉(zhuǎn)碼　出處：《北京郵電大學(xué)》2013年碩士論文　論文類型：學(xué)位論文

【摘要】：近年來(lái),物聯(lián)網(wǎng)和云計(jì)算技術(shù)都在飛速發(fā)展,物聯(lián)網(wǎng)技術(shù)的廣泛應(yīng)用給人們帶來(lái)了巨大的便利,云計(jì)算憑借其強(qiáng)大的存儲(chǔ)和計(jì)算能力,在海量數(shù)據(jù)存儲(chǔ)和分布式計(jì)算領(lǐng)域的應(yīng)用也越來(lái)越多。當(dāng)前物聯(lián)網(wǎng)與云計(jì)算的聯(lián)系越發(fā)緊密,兩者相結(jié)合也是未來(lái)的發(fā)展趨勢(shì),因此如何將云計(jì)算技術(shù)與物聯(lián)網(wǎng)技術(shù)有效的結(jié)合,如何構(gòu)建面向物聯(lián)網(wǎng)的云數(shù)據(jù)中心以及數(shù)據(jù)中心中的計(jì)算、存儲(chǔ)、管理等關(guān)鍵技術(shù)的研究和應(yīng)用便成為了最近幾年研究熱點(diǎn)。 本文深入分析了Hadoop的分布式存儲(chǔ)和MapReduce計(jì)算引擎以及分布式系統(tǒng)管理等技術(shù),對(duì)比了常見云數(shù)據(jù)中心的架構(gòu),包括存儲(chǔ)和計(jì)算模型等,從視頻監(jiān)控系統(tǒng)這一物聯(lián)網(wǎng)具體應(yīng)用場(chǎng)景出發(fā),設(shè)計(jì)并實(shí)現(xiàn)了一種以Hadoop作為核心技術(shù)的面向視頻監(jiān)控領(lǐng)域的云數(shù)據(jù)中心原型系統(tǒng)。論文從數(shù)據(jù)中心硬件基礎(chǔ)平臺(tái)的設(shè)計(jì)和資源整合虛擬化、分布式存儲(chǔ)系統(tǒng)和計(jì)算引擎、視頻監(jiān)控?cái)?shù)據(jù)中心核心服務(wù)等幾個(gè)層面描述了數(shù)據(jù)中心的整體架構(gòu),在對(duì)各層中關(guān)鍵技術(shù)進(jìn)行研究的基礎(chǔ)上,實(shí)現(xiàn)了監(jiān)控視頻分布式采集和管理、海量監(jiān)控視頻文件存儲(chǔ)、分布式視頻文件處理等服務(wù)。相比傳統(tǒng)的視頻監(jiān)控系統(tǒng),基于云計(jì)算的視頻監(jiān)控系統(tǒng)從計(jì)算性能、存儲(chǔ)性能等方面都將有極大的提升。本文對(duì)數(shù)據(jù)中心的核心服務(wù)進(jìn)行測(cè)試和比對(duì),實(shí)驗(yàn)結(jié)果表明本文提出的視頻監(jiān)控?cái)?shù)據(jù)中心系統(tǒng)提高了多路流式視頻數(shù)據(jù)收集的可靠性,顯著降低了離線視頻處理的時(shí)間。
[Abstract]:In recent years, the Internet of things and cloud computing technology are both rapid development, the wide application of Internet of things technology has brought great convenience to people, cloud computing with its powerful storage and computing capabilities. There are more and more applications in the field of mass data storage and distributed computing. At present, the connection between Internet of things and cloud computing is more and more close, and the combination of the two is also the future development trend. Therefore, how to effectively combine cloud computing technology with Internet of things technology, how to build cloud data center for the Internet of things and the computing and storage in the data center. The research and application of key technologies such as management has become a hot topic in recent years. This paper analyzes the distributed storage, MapReduce computing engine and distributed system management technology of Hadoop, and compares the structure of common cloud data center. Including storage and computing models, from the video surveillance system, a specific application of the Internet of things scene. This paper designs and implements a prototype system of cloud data center based on Hadoop, which is oriented to video surveillance field, and designs the hardware platform of data center and the virtualization of resource integration. Distributed storage system, computing engine, core services of video surveillance data center and so on, describe the overall structure of the data center, on the basis of the research on the key technologies in each layer. Compared with the traditional video surveillance system, the video monitoring system based on cloud computing has achieved the performance of video monitoring system based on cloud computing, which realizes distributed video collection and management, mass video file storage, distributed video file processing and other services. Storage performance will be greatly improved. This paper tests and compares the core services of the data center. The experimental results show that the proposed video surveillance data center system improves the reliability of multi-channel video data collection and significantly reduces the time of offline video processing.
【學(xué)位授予單位】：北京郵電大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2013
【分類號(hào)】：TP308;TN948.6

【參考文獻(xiàn)】

相關(guān)碩士學(xué)位論文 前5條

1 李寬;基于HDFS的分布式Namenode節(jié)點(diǎn)模型的研究[D];華南理工大學(xué);2011年

2 郭奕希;基于Hadoop的視頻轉(zhuǎn)碼系統(tǒng)設(shè)計(jì)與實(shí)現(xiàn)[D];華中科技大學(xué);2011年

3 劉f，

本文編號(hào)：1481932