DSQoS-distributed architecture providing QoS in summary warehouses
DOLAP '03 Proceedings of the 6th ACM international workshop on Data warehousing and OLAP
| Hi-index | 0.00 |
Data warehouses are of crucial importance todecision-making in competitive organizations. Thefact that they store enormous quantities of data is achallenge in what concerns performance andscalability, as users request instant answers. None ofthe traditional performance strategies is sufficientlygood to make complex aggregation queries take onlyminutes or seconds. The summary warehouse (SW)achieves such a speedup by storing only general-purpose sampling summaries well-fit for aggregatedexploration analysis.The major limitation of SWs results from thetradeoff between accuracy and speed: smaller, fastersummaries cannot answer less-aggregated queries.In this paper we present the Bag-of-Summariesapproach (BofS) designed to deal effectively with thisproblem: BofS maintains a set of summaries withvaried sizes and chooses the right one to answer aquery with the desired accuracy and best possiblespeedup, based on query granularity considerations.We also present experimental results that show theadvantage of BofS.