本文選題：實(shí)時(shí)遷移　切入點(diǎn)：虛擬機(jī)　出處：《東北大學(xué)》2012年碩士論文

【摘要】：虛擬機(jī)實(shí)時(shí)遷移是虛擬化技術(shù)的一個(gè)重要特征,對于近年來不斷興起的數(shù)據(jù)中心的負(fù)載均衡和災(zāi)難恢復(fù)有非常重大的意義。通過虛擬機(jī)的實(shí)時(shí)遷移,把虛擬機(jī)從源物理主機(jī)硬件平臺遷移到目的物理主機(jī)硬件平臺上,不僅可以快速消除源物理機(jī)上虛擬機(jī)之間的資源競爭,還可以通過合并操作大大減少激活的物理機(jī)數(shù)目,顯著提高計(jì)算的效率。 目前虛擬機(jī)實(shí)時(shí)遷移大多采用迭代預(yù)拷貝方法,該方法在虛擬機(jī)低負(fù)載或網(wǎng)絡(luò)環(huán)境較好的情況下有良好的表現(xiàn),但也具有明顯的局限性。當(dāng)臟頁率過大時(shí),會產(chǎn)生大量內(nèi)存頁重傳,導(dǎo)致遷移期間數(shù)據(jù)傳送總量變大和總遷移時(shí)間變長,影響遷移期間虛擬機(jī)的服務(wù)質(zhì)量；當(dāng)虛擬機(jī)網(wǎng)絡(luò)負(fù)載過重時(shí),會出現(xiàn)臟頁率大于遷移所用網(wǎng)絡(luò)傳輸速率的情況,預(yù)拷貝算法會失效,導(dǎo)致停機(jī)時(shí)間過長。 本文對Xen虛擬機(jī)內(nèi)存實(shí)時(shí)遷移技術(shù)進(jìn)行了深入地研究,針對原有迭代預(yù)拷貝算法在特定環(huán)境下存在的總遷移時(shí)間和停機(jī)時(shí)間過長、遷移期間服務(wù)質(zhì)量降低、算法失效等問題,提出了一種內(nèi)存實(shí)時(shí)壓縮遷移算法。該算法在預(yù)拷貝算法的基礎(chǔ)之上,通過設(shè)計(jì)并實(shí)現(xiàn)內(nèi)存實(shí)時(shí)壓縮算法M2LZO和內(nèi)存頁重復(fù)度檢查算法,在內(nèi)存批量傳送出去前,先使用M2LZO算法對內(nèi)存頁以不同的重復(fù)度作區(qū)分,分別使用適合的壓縮等級進(jìn)行壓縮編碼然后再傳送；在目的物理主機(jī)接收到壓縮數(shù)據(jù)后先用相應(yīng)的譯碼算法進(jìn)行解壓縮,再交給預(yù)拷貝算法恢復(fù)內(nèi)存頁。該算法可以有效減少遷移期間內(nèi)存?zhèn)魉涂偭?縮短總遷移時(shí)間和停機(jī)時(shí)間,而且在網(wǎng)絡(luò)環(huán)境較差的情況下也能表現(xiàn)出良好的適應(yīng)性,一定程度上解決了原算法瓶頸問題。在此基礎(chǔ)上,本文還對Xen遷移管理框架進(jìn)行了優(yōu)化。針對原框架不能自動遷移,設(shè)計(jì)了信息收集器、熱點(diǎn)檢測和遷移策略等機(jī)制；針對原框架存在的性能提升空間,設(shè)計(jì)了遷移算法的多線程并行處理和內(nèi)存空閑頁特殊處理等方法提升遷移性能；針對遷移期間服務(wù)質(zhì)量的保證,對遷移過程中占用資源情況進(jìn)行控制,使遷移本身占用的CPU和網(wǎng)絡(luò)帶寬資源達(dá)到最優(yōu)利用。 本文最后通過實(shí)驗(yàn)對M2LZO算法和優(yōu)化后的遷移管理框架進(jìn)行性能測試。實(shí)驗(yàn)結(jié)果表明,優(yōu)化后的遷移管理框架與優(yōu)化前相比,對不同負(fù)載狀態(tài)下虛擬機(jī)的遷移性能有不同程度的提升。
[Abstract]:Real-time migration of virtual machines is an important feature of virtualization technology, which is of great significance to load balancing and disaster recovery of data centers that have been emerging in recent years.Through the real-time migration of virtual machines, the virtual machines are migrated from the source physical host hardware platform to the destination physical host hardware platform, which not only can eliminate the resource competition between the source physical machine virtual machines quickly.The number of activated physical machines can be greatly reduced by combining operation, and the efficiency of calculation can be improved significantly.At present, the virtual machine real-time migration mostly uses iterative precopy method, this method has good performance in the case of low load virtual machine or good network environment, but also has obvious limitations.When the dirty page rate is too high, a large number of memory pages will be retransmitted, which will result in a large amount of data transfer during migration and a longer total migration time, which will affect the quality of service of virtual machines during migration; when the network of virtual machines is overloaded, the quality of service of virtual machines will be affected.When the dirty page rate is higher than the transfer rate of the migrated network, the precopy algorithm will fail and the downtime will be too long.In this paper, the memory real-time migration technology of Xen virtual machine is deeply studied. Aiming at the problems of the total migration time and downtime of the original iterative precopy algorithm in a specific environment, the quality of service during migration, the failure of the algorithm, etc.A real-time memory compression migration algorithm is proposed.On the basis of precopy algorithm, this algorithm designs and implements the real-time memory compression algorithm M2LZO and the memory page repetition checking algorithm. Before the memory is sent out in batches, the M2LZO algorithm is used to distinguish the memory pages with different repetitions.After receiving the compressed data, the corresponding decoding algorithm is used to decompress it, and then the precopy algorithm is given to restore the memory page.This algorithm can effectively reduce the total amount of memory transfer during the migration period, shorten the total migration time and downtime, and show good adaptability in the case of poor network environment, to a certain extent, solve the bottleneck problem of the original algorithm.On this basis, this paper also optimizes the Xen migration management framework.For the original frame can not be migrated automatically, the information collector, hot spot detection and migration strategy are designed, and the performance improvement space of the original framework is proposed.Multithread parallel processing of migration algorithm and special processing of memory free pages are designed to improve migration performance, aiming at the guarantee of quality of service during migration, the use of resources during migration is controlled.The optimal utilization of CPU and network bandwidth resources is achieved by migration itself.Finally, the M2LZO algorithm and the optimized migration management framework are tested by experiments.The experimental results show that the migration performance of the virtual machine under different load states is improved by the optimized migration management framework.
【學(xué)位授予單位】：東北大學(xué)
【學(xué)位級別】：碩士
【學(xué)位授予年份】：2012
【分類號】：TP302

【參考文獻(xiàn)】

相關(guān)期刊論文 前8條

1 尹國定,衛(wèi)紅;云計(jì)算——實(shí)現(xiàn)概念計(jì)算的方法[J];東南大學(xué)學(xué)報(bào)(自然科學(xué)版);2003年04期

2 譚文輝;;利用VMware實(shí)現(xiàn)數(shù)據(jù)中心服務(wù)器虛擬化[J];艦船電子工程;2008年06期

3 吳亦川,黃奎,鄭健平,孫利民,程偉明;一種自適應(yīng)的健壯TCP/IP報(bào)頭壓縮算法[J];計(jì)算機(jī)研究與發(fā)展;2005年04期

4 劉天田;楊升春;歐中紅;袁由光;;基于消息傳遞并行進(jìn)程遷移技術(shù)的研究與實(shí)現(xiàn)[J];計(jì)算機(jī)科學(xué);2009年04期

5 桂勁松;黃遵國;胡光明;;網(wǎng)絡(luò)服務(wù)器進(jìn)程動態(tài)遷移實(shí)現(xiàn)技術(shù)研究[J];計(jì)算機(jī)工程與科學(xué);2006年04期

6 薛曉輝，，高文;改進(jìn)的算術(shù)編碼[J];計(jì)算機(jī)學(xué)報(bào);1997年11期

7 陳康;鄭緯民;;云計(jì)算:系統(tǒng)實(shí)例與研究現(xiàn)狀[J];軟件學(xué)報(bào);2009年05期

8 項(xiàng)煒;;淺析GCC的代碼優(yōu)化機(jī)制[J];現(xiàn)代經(jīng)濟(jì)信息;2008年04期

本文編號：1691185