摘要
当前B^+树索引结构常采用"溢出页"处理重复出现的键值,易于实现,但存在严重的空间浪费。该文通过改进索引结构的定义,使B^+树能够显式支持重复键值,并给出相应算法。实验证明,采用该方法的P+树能有效减少索引文件的大小,使B^+树满足更多数据库索引的需求。
Overflow page is commonly used to deal with the duplicate keys of a B^＋ tree index, although it is easy to implement, it introduces great space waste. This paper adopts an ahernative approach to make the B^＋tree explicitly support the duplicate keys by revising the index structure, and provides the basic algorithms to implement the idea. Experiment indicates that the revised B^＋ tree can significantly reduce the size of the index file, which broadens the application area of B^＋ tree.
出处
《计算机工程》
CAS
CSCD
北大核心
2009年第5期25-27,共3页
Computer Engineering
基金
上海市国际科技合作基金资助项目"智能交通信息服务技术及移动终端导航系统"(062107037)
上海市国际科技合作基金资助项目"车辆网络感知导航的关键技术与高性能支撑平台"(075 1 07005)
关键词
B^+树
重复键值
溢出页
B^＋ tree
duplicate keys
overflow page
