一种面向海量存储系统的高效元数据集群管理方案

高效的、去中心化的元数据管理方案对大型分布式存储系统的可靠性、可扩展性起至关重要的作用.针对基于Hash划分和基于子树划分的元数据管理方案扩展代价巨大、对集群变动敏感等问题，提出一种基于一致性Hash结构的元数据服务器(metadata server, MDS)集群化方案——CH-MMS(consistent Hash based metadata management schema).CH-MMS在一致性MDS集群上引入虚拟MDS(Virtual MDS)，有效平衡MDS集群负载；将Standby机制与延迟更新策略融合并应用于MDS集群，实现MDS快速失效恢复以及集群变动时零数据迁移量.阐述了CH-MMS的体系结构，介绍了核心数据结构layout-table、虚拟MDS结构、延迟更新机制及相关算法，并对CH-MMS扩展性、容错性作了定性分析.最后通过原型系统和模拟实验说明，CH-MMS具有元数据平衡分布、快速失效恢复、灵活的扩展性以及零结点变动数据迁移量等特点，能满足数据量不断增加的大规模存储集群元数据灵活、高效管理的需求.

A High Performance Management Schema of Metadata Clustering for Large-Scale Data Storage Systems

An efficient, decentralized metadata management schema plays a vital role in large-scale distributed storage systems. The Hash-based partition schema and tree-based partition schema pay huge cost for expansion, and are sensitive to changes in cluster. In response to these problems, CH-MMS(consistent Hash based metadata management schema), is proposed. Virtual MDS (metadata server) is introduced in CH-MMS, and good effect for the cluster’s load balance is proved. Combining the standby mechanism with lazy-update policy, CH-MMS achieves fast failover and zero migration when the cluster changes. Due to its distributed metadata structure, CH-MMS has fast metadata lookup speed. In order to solve the problem that the Hash structure will cause damage to file system hierarchical semantics, a simple and flexible mechanism based on regular expression matching has been introduced. The following work is presented in the paper: 1)Expound the architecture of CH-MMS; 2)Introduce the core data structure of layout-table, virtual MDS and lazy-update policy, and their relevant algorithms; 3)Qualitatively analyze scalability and fault tolerance. The prototype system and simulation show that, CH-MMS is metadata-balancing and has fast failover, flexible expansion and zero migration when cluster changes. CH-MMS can meet the needs of flexible, efficient metadata management of large-scale storage systems with increasing data.