📝 SummarXiv

Abstract

Approximate Nearest Neighbor Search (ANNS) over high-dimensional vectors is a foundational problem in databases, where disk I/O often emerges as the dominant performance bottleneck at scale. Existing graph indexing solutions for disk-based ANNS typically either optimize the storage layout for a given graph or construct the graph independently of the storage layout, thus overlooking their interaction. In this paper, we propose the Block-aware Monotonic Relative Neighborhood Graph (BMRNG), a novel graph structure that jointly considers both geometric distance and storage layout for edge selection, theoretically guaranteeing the existence of I/O monotonic search paths. To address the scalability challenge of BMRNG construction, we further develop a practical and efficient variant, the Block-Aware Monotonic Graph (BAMG), which can be constructed in linear time from a monotonic graph considering the storage layout. BAMG integrates block-aware edge pruning with a decoupled storage design that separates raw vectors from the graph index, thereby maximizing block utilization and minimizing redundant disk reads. Additionally, we design a multi-layer navigation graph for adaptive and efficient query entry, along with a block-first search algorithm that prioritizes intra-block traversal to fully exploit each disk I/O operation. Extensive experiments on real-world datasets demonstrate that BAMG achieves up to 2.1x higher throughput and reduces I/O reads by up to 52% compared to state-of-the-art methods, while maintaining comparable recall.