GrASP: A Generalizable Address-based Semantic Prefetcher for Scalable Transactional and Analytical Workloads

2025-10-13

AI Summary

Existing prefetchers struggle with dynamically growing, privacy-constrained large-scale datasets, requiring frequent fine-tuning and exhibiting poor generalization from limited samples. Method: We propose a context-aware multi-label classification framework that jointly models logical block address increments and query semantics, recasting prefetching as a sequential incremental prediction task. Our approach employs a lightweight, scalable multi-layer LSTM that fuses query embeddings with result encodings, enabling adaptation to data evolution without retraining. Contribution/Results: Evaluated on real-world datasets and industrial benchmarks, our method achieves a 91.4% prefetch hit rate, reduces I/O time by 90.8%, and cuts end-to-end query latency by 57.1%, significantly outperforming state-of-the-art approaches. The core innovations lie in joint incremental-semantic modeling and strong few-shot generalization capability.

Technology Category

• Data Mining & Knowledge Management: Intelligent Query Processing
• Search and Optimization: Learning to Search
• Machine Learning: Learning Preferences or Rankings

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Graph Algorithms and Modeling for the Web: Querying, indexing, and retrieval in Web-related graphs

Abstract

Data prefetching--loading data into the cache before it is requested--is essential for reducing I/O overhead and improving database performance. While traditional prefetchers focus on sequential patterns, recent learning-based approaches, especially those leveraging data semantics, achieve higher accuracy for complex access patterns. However, these methods often struggle with today's dynamic, ever-growing datasets and require frequent, timely fine-tuning. Privacy constraints may also restrict access to complete datasets, necessitating prefetchers that can learn effectively from samples. To address these challenges, we present GrASP, a learning-based prefetcher designed for both analytical and transactional workloads. GrASP enhances prefetching accuracy and scalability by leveraging logical block address deltas and combining query representations with result encodings. It frames prefetching as a context-aware multi-label classification task, using multi-layer LSTMs to predict delta patterns from embedded context. This delta modeling approach enables GrASP to generalize predictions from small samples to larger, dynamic datasets without requiring extensive retraining. Experiments on real-world datasets and industrial benchmarks demonstrate that GrASP generalizes to datasets 250 times larger than the training data, achieving up to 45% higher hit ratios, 60% lower I/O time, and 55% lower end-to-end query execution latency than existing baselines. On average, GrASP attains a 91.4% hit ratio, a 90.8% I/O time reduction, and a 57.1% execution latency reduction.

Problem

• Addressing dynamic dataset scalability in learning-based data prefetchers
• Overcoming limited training data access due to privacy constraints
• Improving prefetching accuracy for both transactional and analytical workloads

Innovation

• Uses logical block address deltas for prefetching
• Combines query representations with result encodings
• Frames prefetching as multi-label classification task