analiza aiqbench
This preprint applies deep learning to database query optimization, using a Transformer and LSTM to estimate how many rows a query will return when it mixes string pattern matching with numeric filters. The reported gains are "up to" 16x lower mean Q-error than PostgreSQL on mixed queries and under 5 ms inference latency, but these are self-reported results from an unreviewed preprint, and the latency and accuracy trade-offs against production optimizers will matter more than the headline numbers. For ML readers, it's a useful example of learned components being pushed into systems infrastructure.
MultiCard: Multiple Attribute Cardinality Estimation via Deep Learning
arXiv:2610.27019v1 Announce Type: cross Abstract: Cardinality estimation is essential for query optimization, but most existing learned estimators focus on a single predicate type, such as string LIKE clauses or numerical range filters. In practice, analytical workloads often combine both types, while the accuracy of string estimators tends to decrease with longer patterns. We introduce MultiCard, a comprehensive learned cardinality estimator that unifies LIKE predicates and numerical filters within a single end-to-end architecture. By combining a Transformer for long-range sparse patterns with an LSTM for short, frequent patterns, MultiCard achieves robust pattern modeling. In parallel, a numerical encoder integrates catalog histograms and structural predicate features, enabling the model to dynamically assess histogram estimation accuracy across different operators and data skew conditions. A final Transformer layer captures interactions among heterogeneous predicates before a multilayer perceptron predicts query cardinalities. We also introduce UPCI, a workload generator that produces length-balanced LIKE patterns, multi-column numerical predicates, and exact cardinality labels for training and evaluation. Experiments on five real-world datasets show that, for string-only workloads, MultiCard achieves the best or near-best Q-error across all datasets compared with state-of-the-art methods. For mixed multi-attribute queries, MultiCard reduces PostgreSQL's mean Q-error by up to 16 times while maintaining an inference latency below 5 ms.
Abstract:Cardinality estimation is essential for query optimization, but most existing learned estimators focus on a single predicate type, such as string LIKE clauses or numerical range filters. In practice, analytical workloads often combine both types, while the accuracy of string estimators tends to decrease with longer patterns. We introduce MultiCard, a comprehensive learned cardinality estimator that unifies LIKE predicates and numerical filters within a single end-to-end architecture. By combining a Transformer for long-range sparse patterns with an LSTM for short, frequent patterns, MultiCard achieves robust pattern modeling. In parallel, a numerical encoder integrates catalog histograms and structural predicate features, enabling the model to dynamically assess histogram estimation accuracy across different operators and data skew conditions. A final Transformer layer captures interactions among heterogeneous predicates before a multilayer perceptron predicts query cardinalities. We also introduce UPCI, a workload generator that produces length-balanced LIKE patterns, multi-column numerical predicates, and exact cardinality labels for training and evaluation. Experiments on five real-world datasets show that, for string-only workloads, MultiCard achieves the best or near-best Q-error across all datasets compared with state-of-the-art methods. For mixed multi-attribute queries, MultiCard reduces PostgreSQL's mean Q-error by up to 16 times while maintaining an inference latency below 5 ms.
Databases (cs.DB); Information Retrieval (cs.IR); Machine Learning (cs.LG)
arXiv:2610.27019 [cs.DB]
(or arXiv:2610.27019v1 [cs.DB] for this version)
https://doi.org/10.48550/arXiv.2610.27019 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Text integral sub licența deschisă a publicației · 299 cuvinte
Subiecte conexe
Fully Inductive Cardinality Estimation
arXiv · 31 iul. 2026
When Does q-error Predict Plan Regret? Three Regimes of Cardinality-Estimation Error
arXiv · 16 iun. 2026
Partially Observed Sparse Graphs: The Unknown Sampling Rate is a Tail Index
arXiv · 23 sept. 2026
Identication and Estimation of Nonsymmetric Determinantal Point Processes
arXiv · 14 oct. 2026