计算机与现代化

• 数据库与数据挖掘 • 上一篇    下一篇

以平衡B链树为数据库索引的并发控制算法

  

  1. 1.上海农林职业技术学院农业信息工程系,上海201699;2.海南医学院医学信息系,海南海口571199
  • 收稿日期:2016-05-03 出版日期:2016-11-15 发布日期:2016-11-23
  • 作者简介:包斌(1973-),男,河南信阳人,上海农林职业技术学院农业信息工程系讲师,硕士,研究方向:数据库及应用; 李亚岗(1971-),男,河南襄城人,海南医学院医学信息系副教授,硕士,研究方向:数据库及应用。

Concurrency Control Algorithms with Balanced Blink-tree Database Index

  1. 1. Department of Agricultural Information Engineering, Shanghai Vocational College of Agriculture and Forestry, Shanghai 201699, China;
    2. Department of Medical Informatics, Hainan Medical University, Haikou 571199, China
  • Received:2016-05-03 Online:2016-11-15 Published:2016-11-23

摘要: 针对B链树作为多版本数据库索引的并发控制机制,提出一种新的B链树结构修改并发控制算法。该算法将B链树结构修改操作划分为几个更小的原子修改操作,这些修改操作可以并发运行并且不会死锁。实验表明,新算法提高了并发度与事务的吞吐量,并保持B链树结构的一致性和平衡性。

关键词: 平衡B链树, 并发, 粒度, 多版本

Abstract: Concerning the concurrency control mechanism for multiversion database index based on Blink-tree, a new Blink-tree concurrency control modification algorithm was proposed. The algorithm divides Blink-tree structure modification into several smaller atomic modifications which run concurrently and deadlock-free. The experimental results show that the new algorithm improves concurrency and transaction throughput, and retains consistency and balance of Blink-tree structure.

Key words: balanced Blink-tree, concurrency, granularity, multiversion

中图分类号: