Beyond TPC-H: Scaling IA2 for Real-World Database Optimization
10 Jan 2026
IA2 sets a new standard for database optimization using the TD3-TD-SWAR model, offering superior efficiency and unseen workload generalization.
Adaptive Action Pruning: Scaling Index Selection for Unseen Workloads
10 Jan 2026
IA2 revolutionizes index selection with rapid training, reducing SQL runtime by 61% via adaptive action pruning and workload modeling
Beyond SWIRL: Scalable Database Indexing for Dynamic Workloads
10 Jan 2026
IA2 outperforms SWIRL and other methods by 15-20%, achieving a 61% runtime improvement through storage-aware RL and rapid training efficiency.
Adaptive Action Pruning: Scaling Databases with Intelligent Action Masking
9 Jan 2026
IA2 outpaces SWIRL and Lan et al. in training efficiency using TD3-TD-SWAR and adaptive action masking for faster SQL index selection.
PostgreSQL & HypoPG: The Experimental Foundation of IA2 Index Selection
7 Jan 2026
IA2 demonstrates significant database optimization on TPC-H benchmarks using PostgreSQL and HypoPG, achieving superior end-to-end runtime gains.
Beyond Preprocessing: Sequential Indexing and Temporal Consistency in IA2
7 Jan 2026
IA2 uses the TD3-TD-SWAR algorithm to adaptively mask irrelevant indexes, ensuring robust performance optimization for both familiar and unseen workloads.
IA2 Preprocessing: Establishing the Foundation for Index Selection
6 Jan 2026
The IA2 preprocessing phase uses a workload model and index candidates enumerator to create accurate state representations and action spaces.
Unseen Workload Optimization: The Two-Phase IA2 Approach
6 Jan 2026
IA2 uses a two-phase framework to generate states and action pools from workloads, enabling RL agents to make sequential index selection decisions.
Minimizing TD Error: Optimizing Computational Efficiency in ISP
24 Dec 2025
The TD3-TD-SWAR model uses a selector network to mask actions, minimizing Temporal Difference (TD) error for faster, high-impact index selection.