计算机科学与探索 ›› 2011, Vol. 5 ›› Issue (7): 595-601.

• 学术研究 • 上一篇    下一篇

面向多虚拟机的分布式存储系统

李 鹤1,2,3, 侯海翔1,2,3, 姜 跃1,2,3, 廖小飞1,2,3, 金 海1,2,3   

  1. 1. 华中科技大学 计算机科学与技术学院, 武汉 430074
    2. 华中科技大学 服务计算技术与系统教育部重点实验室, 武汉 430074
    3. 华中科技大学 集群与网格计算湖北省重点实验室, 武汉 430074

  • 收稿日期:1900-01-01 修回日期:1900-01-01 出版日期:2011-07-01 发布日期:2011-07-01

Distributed Storage for Multiple Virtual Machines

LI He, HOU Haixiang, JIANG Yue, LIAO Xiaofei, JIN Hai   

  1. 1. School of Computer Science and Technology, Huazhong University of Science and Technology, Wuhan 430074, China 2. Key Laboratory of Services Computing Technology and System of Ministry of Education, Huazhong Univer-sity of Science and Technology, Wuhan 430074, China
    3. Key Laboratory of Cluster and Grid Computing of Hubei Province, Huazhong University of Science and Tech-nology, Wuhan 430074, China
  • Received:1900-01-01 Revised:1900-01-01 Online:2011-07-01 Published:2011-07-01

摘要: 通过分析虚拟桌面系统对存储系统的要求, 设计并实现了面向多虚拟机的分布式存储系统—— 虚拟机存储系统(virtual machine storage, VMStore)。分布式技术具有灵活性和可扩展性, 因此采用分布式技术提高多虚拟机存储系统的吞吐率。在元数据管理中采用直接块索引结构以获得更好的快照性能, 并采用去冗余技术减少多虚拟机镜像之间的冗余数据。实验结果显示存储系统具有较好的吞吐能力, 且快照和去冗余的开销对整个系统的影响较小。

关键词: 分布式存储, 虚拟化, 快照, 去冗余

Abstract: By analyzing the application scene of virtualized desktop systems, this paper realizes a distributed storage system for multiple virtual machines: virtual machine storage (VMStore). Because of the flexibility and scalability of the distributed system, using the distributed system can improve the throughput of VMStore. In metadata management, the paper uses the direct block index structure to achieve better performance of snapshot, and the deduplication technology to reduce the redundant data between the images of multiple virtual machines. The experimental results show that VMStore has a good throughput performance, and the overloads of snapshot and deduplication are acceptable.

Key words: distributed storage, virtualization, snapshot, deduplication