Work

Learned Cardinalities: Estimating Correlated Joins with Deep Learning

Year: 2018

Type: other

Abstract: We describe a new deep learning approach to cardinality estimation.MSCN is a multi-set convolutional network, tailored to representing relational query plans, that employs set semantics to capture query features and true cardinalities.MSCN builds on sampling-based estimation, addressing its weaknesses when no sampled tuples qualify a predicate, and in capturing join-crossing correlations.Our evaluation of MSCN using a real-world dataset shows that deep learning significantly enhances the quality of cardinality estimation, which is the core problem in query optimization.

Source: Centrum Wiskunde & Informatica (CWI), the national research institute for mathematics and computer science in the Netherlands

Authors: Andreas Kipf, Thomas Kipf, Bernhard Radke, Viktor Leis, Peter Boncz, Alfons Kemper

Institutions: Technical University of Munich, University of Amsterdam, Centrum Wiskunde & Informatica

Language: English

Topic: Advanced Database Systems and Queries

Subfield: Information Systems

Field: Computer Science

Domain: Physical Sciences

Open Access status: green

Funder: Bundesministerium für Bildung und Forschung

Award: Software Campus (TU München)

OpenAlex ID: W2962868088

Cites: 26

Cited by: 42

Related to: 10