一种负载敏感的OLAP查询结果缓存管理技术

A cache management technology based on load sensitive OLAP query results

OLAP(On Line Analysis Processing)是数据仓库的典型应用,在数据仓库中频繁并发地执行涉及较大数据量的OLAP查询时,其查询处理效率易于逐渐降低。缓存技术是一种有效降低OLAP查询处理延时的方法。在现有的缓存数据存储、淘汰策略等研究工作的基础上,结合OLAP任务的负载特性、OLAP任务的结果集大小等因素对性能的影响,提出了一种负载敏感的OLAP查询缓存管理技术Workload-LRU,并实现了一个ROLAP(Relational OLAP)原型系统。实验证明,Workload-LRU技术获得了较好的性能提升效果。

OLAP is a typical application of data warehouse. High concurrent data-intensive OLAP tasks often tend to incur large resource consumption and make the data warehouse system to be inefficient. Caching technology has been recognized as an effective method to accelerate the execution of OLAP tasks. In this paper,a workload sensitive OLAP query cache management technology,named Workload-LRU,has been proposed by carefully incorporate OLAP task＇s performance metrics,such as workload characteristics,size of results set and so on. To evaluate the effectiveness of Workload-LRU,we implement it into a ROLAP prototype system,and the experimental results show that the proposed Workload-LRU technology is superior to the representative LRU based techniques.