摘要： 关系数据库中的索引技术可以快速判断记录重复，但对于频繁更新的海量数据库，维护索引的时间与资源开销较大。针对交通量数据包及其海量数据库的特点，提出一个交通量实时包的时序区间模型，给出并证明了一个基于区间记录的快速判重算法，分析了算法的复杂度，探讨了改进算法的方法。该算法具有复杂度与数据库大小无关、高效、易于实现等特点。

关键词: 海量数据库, 重复, 公路交通量, 实时包

Abstract: The index technique in relational database can judge rapidly a duplicate record. But the cost of time and resource is more for maintaining the index in massive database when its records updates frequently. Considering the characteristic of traffic packet and traffic massive database, this paper puts forward a time order interval model for traffic real-time packet, gives and proves an algorithm for rapidly judging a duplicate packet based on interval records, analyses the algorithm complexity, and discusses several improved methods. The algorithm introduced here has the characteristic that the complexity is independent on the database size, the efficiency is high, and can be realized easily.

Key words: massive database, duplicate, highway traffic, real-time packet

中图分类号: 

• TP311.13