A GPU-Accelerated Framework for Multi-Attribute Range Filtered Approximate Nearest Neighbor Search

2026-04-21

AI Summary

This work addresses the limitations of existing range-filtered approximate nearest neighbor search (ANNS) methods, which suffer from index bloat, high construction overhead, and CPU-only execution, hindering efficient query processing. To overcome these challenges, we propose a lightweight GMG index structure coupled with a hardware-aware execution pipeline. Our approach achieves linear storage complexity through cell-based partitioning and local graph construction, introduces a cluster-guided query reordering mechanism to effectively harness GPU parallelism, and incorporates a cell-oriented out-of-core pipeline to circumvent GPU memory constraints. To our knowledge, this is the first solution enabling GPU acceleration for multi-attribute range-filtered ANNS. Experimental results demonstrate that our method reduces index size by 4.4× while achieving a 119.8× throughput improvement over the state-of-the-art, all while maintaining high recall.

Technology Category

• Machine Learning: Hardware-aware ML
• Search and Optimization: Distributed Search
• Data Mining & Knowledge Management: Intelligent Query Processing

Application Category

• Graph Algorithms and Modeling for the Web: Querying, indexing, and retrieval in Web-related graphs
• Search and Retrieval-Augmented AI: Efficiency and scalability of Web search engines
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

Range-filtered approximate nearest neighbor search (RFANNS) is increasingly critical for modern vector databases. However, existing solutions suffer from severe index inflation and construction overhead. Furthermore, they rely exclusively on CPUs for the heavy indexing and query processing, failing to leverage the powerful computational capabilities of GPUs. In this paper, we present Garfield, a GPU-accelerated framework for multi-attribute range filtered ANNS that overcomes these bottlenecks through designing a lightweight index structure and hardware-aware execution pipeline. Garfield introduces the GMG index, which partitions data into cells and builds local graph indexes. By adding a constant number of cross-cell edges, it guarantees linear storage and indexing overhead. For queries, Garfield utilizes a cluster-guided ordering strategy that reorders query-relevant cells, enabling a highly efficient cell-by-cell traversal on the GPU that aggressively reuses candidates as entry points across cells. To handle datasets exceeding GPU memory, Garfield features a cell-oriented out-of-core pipeline. It dynamically schedules cells to minimize the number of active queries per batch and overlaps GPU computation with CPU-to-GPU index streaming. Extensive evaluations demonstrate that Garfield reduces index size by 4.4x, while delivering 119.8x higher throughput than state-of-the-art RFANNS methods.

Problem

• range-filtered approximate nearest neighbor search
• index inflation
• GPU acceleration
• vector databases
• multi-attribute filtering

Innovation

• GPU acceleration
• range-filtered ANNS
• lightweight index
• out-of-core processing
• cell-based traversal