This paper designs a feedback-oriented vector database with query structure-aware embed-dings to support effective similarity search, and incorporates multi-layered histories as references to enrich the feedback, and enhances the prompt optimization work-flow by utilizing multi-dimensional historical feedback to drive continuous self-evolution.
LLM-QOpt++ is presented, a novel hybrid, confidence-aware query optimization framework that unifies traditional CBO estimation, machine learning–based cost prediction, and large language model (LLM) reasoning within a single adaptive pipeline.
A KL-divergence-driven model fine-tuning and plan updating strategy that dynamically adapts to workload changes is introduced, demonstrating improved scalability and robustness for industrial-scale PQO.
Song-Song Mo, Quan-Qing Xu, Xu-Chen Ding et al.· Proceedings of the VLDB Endo...· 0 citations
DBRepro is presented, an automated end-to-end framework that formulates database generation as a constrained distribution synthesis problem that reproduces the execution performance of complex slow queries with high fidelity on a nearly 1 TB real-world dataset.
Zhao-Yang Zhang, Shuang Liu, Deng-Feng Xu et al.· Proceedings of the 41st IEEE...· 0 citations
A combined approach is proposed that fine-tunes the model while taking KV cache concatenation into account and selectively recomputes a subset of the KV caches and demonstrates improved accuracy for long-context inputs.
F. Tachibana, Daisuke Miyashita, Jun Deguchi· 0 citations
This work adapts Balsa to DuckDB, a vectorized analytical database that lacks both plan hinting and statement-level timeout support, and introduces lightweight portability mechanisms based on SQL-level plan forcing and timeout management for safe exploration.
SKYPQ is demonstrated, a system that employs a novel, dominance-based data reduction method for package queries under updates that maintains a K-skyband index, a small, correctness-preserving subset of candidates, and applies a resolve checker to skip re-optimization when updates cannot affect the optimal package.
Vasileios Vittis, Azza Abouzied, Peter J. Haas et al.· Proceedings of the VLDB Endo...· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
