DGAI: Decoupled On-disk Graph-based ANN Index For Efficient Updates And Queries

Abstract

On-disk graph-based indexes are favored for billion-scale Approximate Nearest Neighbor Search (ANNS) due to their high performance and cost-efficiency. However, existing systems typically rely on a coupled storage architecture that co-locates vectors and graph topology, which introduces substantial redundant I/O during index updates, thereby degrading usability in dynamic workloads. In this paper, we propose a decoupled storage architecture that physically separates heavy vectors from the lightweight graph topology. This design substantially improves update performance by reducing redundant I/O during updates. However, it introduces I/O amplification during ANNS, leading to degraded query efficiency.To improve query performance within the update-friendly architecture, we propose two techniques co-designed with the decoupled storage. We develop a similarity-aware dynamic layout that optimizes data placement online so that redundantly fetched data can be reused in subsequent search steps

Stats

• citations0
• S2 citations—
• github stars0
• HF likes0
• heat score0.00
• arxiv keylou2025dgai

Related papers

• In-place Updates Of A Graph Index For Streaming Approximate Nearest Neighbor Search (2025)0.00
• BAMG: A Block-aware Monotonic Graph Index For Disk-based Approximate Nearest Neighbor Search (2025)0.00
• Freshdiskann: A Fast And Accurate Graph-based ANN Index For Streaming Similarity Search (2021)0.00
• Diskann++: Efficient Page-based Search Over Isomorphic Mapped Graph Index Using Query-sensitivity Entry Vertex (2023)0.00
• Frequency-aware Graph Construction And Search For Dynamic Vector Databases (2025)0.00
• MCGI: Manifold-consistent Graph Indexing For Billion-scale Disk-resident Vector Search (2026)0.00
• Ood-diskann: Efficient And Scalable Graph ANNS For Out-of-distribution Queries (2022)0.00
• Breaking The Storage-compute Bottleneck In Billion-scale ANNS: A Gpu-driven Asynchronous I/O Framework (2025)0.00