基于物化策略的异构数据集成的动态查询调度

主要是引入了物化策略来优化异构数据集成调度算法,从总体上介绍了异构数据集成系统的查询处理流程.为了解决传统查询调度存在的几个问题:成本估计不正确,运行时可用的内存不足和数据传输率无法预测,提出了满足内存限制的一种查询调度的动态策略,主要包括调度条件、监察临界值、物化策略.最后根据动态调度优化对子查询结果进行整合,通过实验证明本文中动态查询调度的优越性.

Dynamic Query Scheduling for Heterogeneous Data Integration System

The main purpose of this paper is to optimize the heterogeneous data integration scheduling algorithm by materialization strategies,the query processing flow is introduced from the whole.In order to solve the traditional questions of query scheduling: the cost estimates may be inaccurate,the memory available at run-time may be insufficient,or the data delivery rate can be unpredictable,we propose to dynamically schedule queries to deal with irregular data delivery rate and gracefully adapt to the available memory.It includes schedulability conditions,detecting critical,materialization strategies.At last,integration of subquery result is done according to the dynamic optimization scheduling and the advantage of the dynamic query scheduling is proved by experiment.