Graph-based high-dimensional vector indices have become a mainstream solution for large-scale approximate nearest neighbor search (ANNS). However, their substantial memory footprint often requires storage on secondary devices, where frequent on-demand loading of graph and vector data makes I/O the dominant bottleneck, accounting for over 90% of query latency.
GoVector is an I/O-efficient caching strategy designed for disk-based graph indices. It builds on top of DiskANN and combines:
- a static cache that stores entry points and frequently accessed neighbors, and
- a dynamic cache that adaptively captures nodes with high spatial locality during query processing.
In addition, GoVector optimizes the disk layout by colocating similar vectors on the same or adjacent pages, improving locality and reducing I/O overhead.
Experimental results on multiple public datasets show that GoVector significantly improves performance. At 90% recall, it reduces I/O operations by 46% on average, increases query throughput by 1.73×, and reduces query latency by 42% compared to state-of-the-art disk-based graph indexing systems.
If you find this work useful, please cite GoVector and its base project DiskANN:
@misc{zhou2025govectorioefficientcachingstrategy,
title={GoVector: An I/O-Efficient Caching Strategy for High-Dimensional Vector Nearest Neighbor Search},
author={Yijie Zhou and Shengyuan Lin and Shufeng Gong and Song Yu and Shuhao Fan and Yanfeng Zhang and Ge Yu},
year={2025},
eprint={2508.15694},
archivePrefix={arXiv},
primaryClass={cs.DB},
url={https://arxiv.org/abs/2508.15694},
}
@misc{diskann-github,
author = {Simhadri, Harsha Vardhan and Krishnaswamy, Ravishankar and Srinivasa, Gopal and Subramanya, Suhas Jayaram and Antonijevic, Andrija and Pryce, Dax and Kaczynski, David and Williams, Shane and Gollapudi, Siddarth and Sivashankar, Varun and Karia, Neel and Singh, Aditi and Jaiswal, Shikhar and Mahapatro, Neelam and Adams, Philip and Tower, Bryan and Patel, Yash},
title = {{DiskANN: Graph-structured Indices for Scalable, Fast, Fresh and Filtered Approximate Nearest Neighbor Search}},
url = {https://github.com/Microsoft/DiskANN},
version = {0.6.1},
year = {2023}
}