Workload-Based Wavelet Synopses
msra(2003)
摘要
This paper introduces workload-based wavelet synopses, which exploit query workload information to signican tly boost accuracy in approximate query processing. We show that wavelet synopses can adapt eectiv ely to workload information, and that they have signican t advantages over previous approaches. An important aspect of our approach is optimizing synopses constructions toward error metrics dened by workload information, rather than based on some uniform metrics. We present an adaptive greedy algorithm which is simple and ecien t. It is run-time competitive to previous, non-workload based algorithms, and constructs workload-based wavelet synopses that are signican tly more accurate than previous synopses. The algorithm also obtains improved accuracy for non-workload case when the error metric is the mean relative error. We also present a self-tuning algorithm that adapts the workload-based synopses to changes in the workload. All algorithms are extended to workload-based multidimensional wavelet synopses with im- proved performance over previous algorithms. Experimental results demonstrate the eectiv eness of workload-based wavelet synopses for dieren t types of data sets and query workloads, and show signi- cant improvement in accuracy even with very small training sets.
更多查看译文
关键词
relative error,greedy algorithm
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络