中国科学院软件研究所机构知识库
Advanced  
ISCAS OpenIR  > 综合信息系统技术国家级重点实验室  > 学位论文
Subject: 计算机应用::计算机应用其他学科
Title:
虚拟机跨数据中心动态迁移的内存去重技术研究
Author: 郑冕
Issued Date: 2015-05-27
Supervisor: 胡晓惠
Major: 计算机应用技术
Degree Grantor: 中国科学院研究生院
Place of Degree Grantor: 北京
Degree Level: 硕士
Keyword: 虚拟机 ; 动态迁移 ; 数据中心 ; 数据去重 ; 模板页面
Abstract: 随着云计算产业的发展,其核心技术之一——虚拟化技术在数据中心中得到了广泛的应用。虚拟机的动态迁移作为虚拟化技术的一项重要特性,在数据中心的管理中有着重要作用。而虚拟机跨数据中心动态迁移可以在更大范围的联合云或者混合云环境中帮助进行资源整合,动态任务调度以及负载均衡。随着越来越多的数据中心投入运行,这种跨数据中心进行虚拟机动态迁移的需求将会日益增加。 首先,本文针对虚拟机跨数据中心动态迁移中内存状态变化速度较快与广域网传输延迟较大之间的矛盾,结合对大量虚拟机内存数据的分析以及数据去重方法的思想,提出了一种基于模板的内存去重方法(template-based memory deduplication method, TMDM)。该方法通过设置模板页面阈值,只对重复出现次数达到一定要求的内存页面进行索引。这样既考虑到了大范围内存数据的特点,保证了高去重率,又控制住了去重需比对的内存页面集合的规模。该方法在内存状态的首轮迭代和后续迭代拷贝过程中均能有效降低传输数据量。一系列虚拟机的动态迁移实验表明,TMDM方法在迁移性能的各个维度均明显优于具有代表性的Shrinker和QEMU/KVM默认的动态迁移方法。 其次,针对TMDM方法将模板页面内容置于目的端物理节点内存上的设计,本文分析得出在大中型数据中心场景下,由于模板页面规模增大而使得TMDM方法的使用存在问题。由此提出一种基于分布式哈希表的虚拟机动态迁移架构(template-based virtual machines live migration between data centers using distributed hash table, TMDHT),该架构在目的端数据中心中构建一个采用分布式哈希表(distributed hash table,DHT)思想的环状结构,并将模板页面集合分片,分布式地存储于目的端数据中心内的多个节点上。模板页面的恢复基于TMDHT架构的路由机制。实验结果表明,这样的虚拟机动态迁移架构在保证明显的迁移性能提升的基础上,极大地提高了TMDM方法的可扩展性和可用性。 最后,本文结合对QEMU/KVM开源平台的深入研究,以此平台为基础,实现了一个利用TMDM方法进行内存去重的动态迁移原型系统,并通过实现TMDHT架构对其进行改进。原型系统的实现是TMDM方法、TMDHT架构分别与其他方法进行性能对比实验的基础。
English Abstract: Along with the development of cloud computing industry, the virtualization technology, as one of the core technologies of cloud computing, has been widely implemented in data centers. The virtual machines live migration is an important feature of virtualization, and plays an important role in the management of data centers. Inter-data centers live migration of virtual machines enables resource integration, task scheduling and load balancing in a larger scale. Along with the growing number of data centers, the inter-data center virtual machines live migration will be more and more important. First of all, this paper focuses on the contradiction between the fast changing memory state and the low transmission speed over WAN during virtual machines live migration between data centers. In combination with the memory data analysis and the data deduplication method, a template-based memory deduplication method (TMDM) is proposed. This method only indexes those memory pages which occur more times than a preset threshold. It takes a wide range of memory data into account to ensure the high deduplication rate, and controls the scale of template pages at the same time. The TMDM can reduce the amount of transferred data in the first round and the subsequent rounds of memory copy. The experimental results show that the TMDM outperforms Shrinker and the default method of QEMU/KVM. Second, this paper optimizes the design of pre-storing all the template pages in memory of destination host in the TMDM. As the data amount of template pages will be more than a single host can hold. The template-based virtual machines live migration between data centers using distributed hash table (TMDHT) is proposed. This architecture exploits the structure of distributed hash table (DHT). In TMDHT, template pages are sliced and stored distributedly in the destination data center. The destination host retrieves pages following routing rules. The experimental results show that TMDHT greatly increases the scalability and availability of TMDM, while keeping the high migration performance. Finally, in combination with the in-depth study of QEMU/KVM open source platform, this paper implements a virtual machines live migration prototype. This prototype exploits TMDM to do data deduplication and then improves with TMDHT. The implementation of prototypes is the foundation of migration performance comparative experiments.
Content Type: 学位论文
URI: http://ir.iscas.ac.cn/handle/311060/17079
Appears in Collections:综合信息系统技术国家级重点实验室 _学位论文

Files in This Item:
File Name/ File Size Content Type Version Access License
毕业论文完整版_郑冕.pdf(4076KB)----限制开放 联系获取全文

Recommended Citation:
郑冕. 虚拟机跨数据中心动态迁移的内存去重技术研究[D]. 北京. 中国科学院研究生院. 2015-05-27.
Service
Recommend this item
Sava as my favorate item
Show this item's statistics
Export Endnote File
Google Scholar
Similar articles in Google Scholar
[郑冕]'s Articles
CSDL cross search
Similar articles in CSDL Cross Search
[郑冕]‘s Articles
Related Copyright Policies
Null
Social Bookmarking
Add to CiteULike Add to Connotea Add to Del.icio.us Add to Digg Add to Reddit
所有评论 (0)
暂无评论
 
评注功能仅针对注册用户开放,请您登录
您对该条目有什么异议,请填写以下表单,管理员会尽快联系您。
内 容:
Email:  *
单位:
验证码:   刷新
您在IR的使用过程中有什么好的想法或者建议可以反馈给我们。
标 题:
 *
内 容:
Email:  *
验证码:   刷新

Items in IR are protected by copyright, with all rights reserved, unless otherwise indicated.

 

 

Valid XHTML 1.0!
Copyright © 2007-2020  中国科学院软件研究所 - Feedback
Powered by CSpace