摘要
基于一种广泛接受的内存数据库系统基本概念及内存数据库中的SB 树索引结构 ,提出了更方便关系代数操作的SB 树 .在此基础上 ,提出了优化的关系代数操作的实现算法 .根据这些算法和关系代数等价定理 ,给出了对关系代数查询树进行逻辑优化的规则 ,并证明了逻辑优化的正确性 .最后给出了代价评估模型 ,实验表明提出的内存数据库查询优化算法相对于传统的优化算法具有更好的时间及空间性能 .
Based on a concept of main memory database system (MMDBS) and the SB tree indexing structure, this paper proposed a SB tree indexing structure which was more suitable for relational algebra operations in an MMDBS. The implementation algorithms for the optimized relational algebra operation were introduced. With these algorithms and the equivalence theory, the several rules to logically optimize the accessing trees were presented and the correctness for the logical optimization was proved. A model was given to evaluate the cost. Compared with the common query optimization strategies, the experiment results showed that the proposed query optimization algorithms for the main memory database system had better performance in the cost of time and space.
出处
《华中科技大学学报（自然科学版）》
EI
CAS
CSCD
北大核心
2003年第4期21-23,29,共4页
Journal of Huazhong University of Science and Technology(Natural Science Edition)
基金
国家自然科学基金资助项目 (6 0 2 0 30 1 7)
