Abstract: Efficient and scalable data management becomes increasingly important in large-scale distributed storage systems. A key enabling technique is a flexible, balancing and scalable data object placement and location scheme that automatically adapts to the dynamic change of storage nodes. An algorithm for data placement is proposed, which is based on laws of large number. The basic algorithm promises probabilistically even data distribution and minimizing data movement when the number of storage nodes dynamically changes.

Key words: distributed storage, data placement, self-adaptive

摘要： 数据量的快速增长，使得研究能够自动适应存储节点动态变化的数据分布方法成为分布式文件系统领域的难点和热点。基于贝努利大数定律提出一种自适应存储节点规模动态变化的数据分布算法，通过理论分析和实验证明，该算法能够实现在节点规模动态变化过程中数据分布的均衡性，并能保证迁移的数据量从统计意义上最优。

关键词: 分布式存储, 数据分布, 自适应

CLC Number: 

• TP311.13