基于缓冲池描述的DBMS分层排队网络模型

Buffer poolbased DBMS modeling with layered queuing networks

建立数据库管理系统(DBMS)的性能模型是DBMS性能管理的一项核心任务,针对缓冲池对DBMS性能的影响,以缓冲池的工作原理为基础,提出基于缓冲池描述的DBMS分层排队网络模型(LQNM)。该模型把DBMS抽象为3个软件任务(查询进程、逻辑读、物理读)和4个硬件任务(THINK、CPU、内存、硬盘),以负载特征参数为输入,预测负载的响应时间。实验结果表明,该模型比常用的排队网络模型的预测准确性更高,预测平均误差小于15%。

Modeling database management systems （DBMSs） is the key to managing the performance of DBMSs. Buffer pools play an important role in defining the performance of DBMSs. According to the mechanism of Buffer Pools, a layered queuing network model （LQNM） is proposed to model the performance of DBMSs. The model consists of three software tasks, namely, query agent, logic read and physical read, and four hardware tasks, namely, THINK, CPU, RAM and Hard Disk. With the workload parameters as inputs, the model predicts the response time of the workload. Experiments suggest that the proposed model is more accurate than the widely used queuing network model and its prediction error is less than 15 %.