大数据背景下,分布式数据库在使用过程中经常需要扩容,在扩容过程中,各存储节点的哈希值需要重新计算,数据对象也需要大量迁移数据。本文通过预留子分区识别位、数据库扩容过程中物理存储节点编码“高位不变,低位置1”等技术手段实现数据库的高效扩展。对比实验标明:该方法避免了分布式数据库在扩容时的已有存储节点哈希值的重新计算工作,减少了数据对象的数据迁移量,提高了分布式数据库的扩展效率。
Under the big date background, distributed database became bigger during use. In the process of distributed database expansion, storage node’s hash value would be recomputed, and the data object would be migration, so a large amount of data will be transferred. An efficient expansion method for distributed database was adopted, by reserved identification bit and storage node’s code “High Bit Constant, Low Bit Set”. Contrast experiment shows that this method can avoid recomputed hash value of existing storage nodes as well as reduce data migration, and improve the efficiency of the distributed database expansion.
Karger, D., Lehman, E., Leighton, T., et al. (1997) Consistent Hashing and Random Trees: Distributed Caching Protocols for Relieving Hot Spots on the World Wide Web. Proceedings of the Twenty-Ninth Annual ACM Symposium on the Theory of Computing, El Paso, Texas, 4-6 May 1997, 654-663.