以大型数据库应用为背景,基于多Agent技术,构建一个分布式数据库访问平台。研究分布式环境下的语义缓存技术,提出一种Agent平台下的智能预取算法,以及一种改进的黑板模型,用于实现多Agent间的通信。在大型数据库系统上的性能测试结果表明,该方案对海量数据的统计查询性能有显著提高。
With the application of large-scale database in the background, a distributed database accessing platform based on multi-Agent is constructed. Distributed semantic caching is studied, and an intelligent pre-fctching algorithm in Agent-based platform is proposed. A new blackboard model is designed to realize the communication between Agents. The test on large-scale database system shows that the statistic query performance has been improved markedly.