分布式存储系统可用性与一致性分析-计算机应用技术专业论文.docx

华中科技 大学硕士 学 华 中 科 技 大 学 硕 士 学 位 论 文 II II Abstract Consistency, availability, partition tolerance and scalability are mainly difficult problems in the design of distribute storage system. Now many distribute file systems and storage systems are designed to enhance the performance of systems. Our system is designed to build high-availability storage service on inexpensive commodity devices. This system is a distribute key value storage system which use external high-availability system components Zookeeper for leader election. And System use variant Paxos distribute protocol to ensure the strong consistency of replicas. The Node of system can do automatic failover and can ensure no data lose after node recovery. In order to improve system scalability and availability, System support three copy of each data and partition of data. System provide optional consistency of services, including strong consistency and monotonous consistency. The implementation of system show that the overhead of our distribute copy protocol is very small. We use many optimum measures in system implementation, include the parallelization of the network and disk, parallelization of multi requests, serialization of write operations and so on. The write performance of the system have reach the bottleneck of the disk. Also system can provide quite strong consistency service and better monotonous consistency service. Our practice shows that Paxos and other similar quorum distributed protocol are good choice of high-availability system implementation. Key words: Distribute Storeage System, High-availability, Consistency, Scalability, Partition tolerance III III 目 录 摘 要 I HYPERLINK \l _bookmark0 Abstract II HYPERLINK \l _bookmark1 1 绪论 HYPERLINK \l _bookmark2 1.1 研究背景与意义 (1) HYPERLINK \l _bookmark3 1.2 分布式存储系统研究现状 (1) HYPERLINK \l _bookmark4 1.3 本文研究内容以及论文结构 (3) HYPERLINK \l _bookmark5 2 分布式存储系统分布式副本协议分析 HYPERLINK \l _bookmark6 2.1 分布式副本协议分析 (5) HYPERLINK \l _bookmark7 2.2 系统使用的分布式副本方案 (