基于Agent的分布式数据库查询优化研究

目前分布式数据库查询优化器基于代价模型,总代价=CPU代价+I/O代价+通信代价,在远程通信网或数据传输率较低的系统中,通信代价应作为首要的优化目标来考虑;提出了一种基于Multi-Agent技术的查询模型(QMAS);同时为了保证系统组内、组间Agent之间的高效通信,采用了树型拓扑结构(TTMAS)的通信模型,系统内各Agent使用通信原语高效通信、协作,且满足Agent间的通信路由最优,从而保证了QMAS的查询效率.

Agent-Based Query Optimization in Distributed Database

Currently distributed database query optimizer based on the cost model, the total cost=CPU cost + I/O+cost of communication, in the telecommunications network or in the data transfer rate low system, communications costs should serve as the primary goal of optimizing to consider; then the author proposed one kind semantic query model based on multi-agent technology (QMAS), and taking this model as the foundation, he has established one kind parallel database semantics query method based on the agent, and to ensure the high efficiency of communications between the agents of the systems groups, he has also adopted a communication model by Tree Topology Multi-Agents System (TTMAS), every agent in this system communicates and collaborates at high efficient by using communication primitive, thereby ensuring the query efficiency in QMAS.