本文選題：云計(jì)算　切入點(diǎn)：基礎(chǔ)設(shè)施云　出處：《北京郵電大學(xué)》2012年碩士論文　論文類型：學(xué)位論文

【摘要】：云計(jì)算是一種互聯(lián)網(wǎng)計(jì)算新模式,因?yàn)槠渖虡I(yè)模式代表了未來的趨勢(shì),近年來受到了工業(yè)界和學(xué)術(shù)界的廣泛關(guān)注,并且已經(jīng)逐步應(yīng)用(?)起來。其中,基礎(chǔ)設(shè)施云是云計(jì)算三種交付模型(即：軟件云、平臺(tái)云和基礎(chǔ)設(shè)施云)的最基本形式,也是其它兩種云建立的基礎(chǔ)。而基礎(chǔ)設(shè)施云主要是以虛擬化運(yùn)行環(huán)境為服務(wù)載體,為用戶提供彈性的虛擬計(jì)算資源。 本文以云計(jì)算中的虛擬化運(yùn)行環(huán)境為研究對(duì)象,首先調(diào)研并對(duì)比分析了國(guó)內(nèi)外主流的基礎(chǔ)設(shè)施云系統(tǒng)及其虛擬化運(yùn)行環(huán)境的特點(diǎn)。在此叢礎(chǔ)上,研究、實(shí)現(xiàn)了云計(jì)算平臺(tái)中的虛擬化運(yùn)行環(huán)境系統(tǒng)。系統(tǒng)包含資源管理、節(jié)點(diǎn)調(diào)度、虛擬機(jī)生命周期管理和虛擬機(jī)監(jiān)控四個(gè)模塊。資源管理模塊實(shí)現(xiàn)管理和虛擬化異構(gòu)、分布的物理資源(CPU、內(nèi)存、硬盤等),并實(shí)現(xiàn)這些物理資源的按需使用。節(jié)點(diǎn)調(diào)度模塊實(shí)現(xiàn)根據(jù)調(diào)度策略選擇最優(yōu)節(jié)點(diǎn)控制器。虛擬機(jī)生命周期管理模塊實(shí)現(xiàn)在虛擬機(jī)生命周期內(nèi)對(duì)虛擬機(jī)的管理。虛擬機(jī)監(jiān)控模塊實(shí)現(xiàn)對(duì)云計(jì)算平臺(tái)內(nèi)所有節(jié)點(diǎn)控制器上的虛擬機(jī)的實(shí)時(shí)監(jiān)控。四個(gè)模塊相互作用于配合,從而使得系統(tǒng)可以充分、合理的使用和共享系統(tǒng)內(nèi)的資源,使之得到最大化利用。 最后,通過實(shí)驗(yàn)對(duì)比分析系統(tǒng)對(duì)虛擬資源管理的引入代價(jià),評(píng)估不同的虛擬機(jī)管理器、虛擬化方式在CPU、內(nèi)存、I/O、網(wǎng)絡(luò)四個(gè)方面的虛擬化性能差異,并對(duì)個(gè)人和企業(yè)在今后使用云計(jì)算平臺(tái)中的虛擬運(yùn)行環(huán)境,如何根據(jù)工作負(fù)載的不同選擇合適的虛擬機(jī)管理器及虛擬化方式給出合理化建議,從而使得資源利用最大化。
[Abstract]:Cloud computing is a new model of Internet computing, because its business model represents the future trend, has been widely concerned by industry and academia in recent years, and has been gradually applied. Where the infrastructure cloud is the most basic form of the three delivery models of cloud computing, namely, the software cloud, the platform cloud, and the infrastructure cloud, It is also the foundation of the other two kinds of clouds, and the infrastructure cloud is mainly based on the virtualized running environment as the service carrier to provide users with flexible virtual computing resources. In this paper, the virtualization running environment in cloud computing is taken as the research object. Firstly, the characteristics of the mainstream infrastructure cloud system and its virtualized running environment at home and abroad are investigated and compared. The system includes four modules: resource management, node scheduling, virtual machine lifecycle management and virtual machine monitoring. Distributed physical resources CPU, memory, The node scheduling module realizes the selection of the optimal node controller according to the scheduling strategy. The virtual machine lifecycle management module implements the implementation of the virtual machine in the virtual machine life cycle. Management. Virtual machine monitoring module realizes real-time monitoring of virtual machines on all node controllers in cloud computing platform. The four modules interact with each other. So that the system can fully and reasonably use and share the resources in the system to maximize its utilization. Finally, by comparing and analyzing the cost of virtual resource management, we evaluate the virtualization performance of different virtual machine managers, virtualization methods in CPU, memory I / O, and network. In order to maximize the utilization of resources, the author gives some reasonable suggestions on how to select the appropriate virtual machine manager and virtualization mode according to the different workload in the virtual running environment of individuals and enterprises using cloud computing platform in the future.
【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】：2012
【分類號(hào)】：TP3

【引證文獻(xiàn)】

相關(guān)碩士學(xué)位論文 前1條

1 戴琪;云計(jì)算的自動(dòng)部署服務(wù)在銀行Power系統(tǒng)上的應(yīng)用研究[D];上海交通大學(xué);2013年

，

本文編號(hào)：1610029