arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2006.08109cs.DBcs.LG
收藏

NeuroCard:所有表的一个基数估计

NeuroCard: One Cardinality Estimator for All Tables

UC Berkeley(加州大学伯克利分校);Covariant(协变公司)

Zongheng Yang, Amog Kamsetty, Sifei Luan, Eric Liang, Yan Duan, Xi Chen, Ion Stoica

摘要:

查询优化器依赖于准确的基数估计来生成良好的执行计划。尽管经过几十年的研究,现有的基数估计器对于复杂查询是不准确的,这是由于进行了有损建模假设并且没有捕获表间的相关性。在这项工作中,我们证明了在没有任何独立性假设的情况下,可以学习数据库中所有表的相关性。我们提出了NeuroCard,一个连接基数估计器,在整个数据库中构建一个单一的神经密度估计器。利用联合采样和现代深度自回归模型,NeuroCard在其概率建模中没有表间或列间独立性假设。NeuroCard实现了比现有最佳方法高几个数量级的准确性(在JOB-light上最大误差为8.5 $\times$的最新结果),可扩展到数十个表,同时空间紧凑(几MB),构建或更新效率高(秒到分钟)。

英文摘要:

Query optimizers rely on accurate cardinality estimates to produce good execution plans. Despite decades of research, existing cardinality estimators are inaccurate for complex queries, due to making lossy modeling assumptions and not capturing inter-table correlations. In this work, we show that it is possible to learn the correlations across all tables in a database without any independence assumptions. We present NeuroCard, a join cardinality estimator that builds a single neural density estimator over an entire database. Leveraging join sampling and modern deep autoregressive models, NeuroCard makes no inter-table or inter-column independence assumptions in its probabilistic modeling. NeuroCard achieves orders of magnitude higher accuracy than the best prior methods (a new state-of-the-art result of 8.5$\times$ maximum error on JOB-light), scales to dozens of tables, while being compact in space (several MBs) and efficient to construct or update (seconds to minutes).

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑