Skip to content

README.md

Latest commit

 

History

History
100 lines (66 loc) · 3.19 KB

README.md

File metadata and controls

100 lines (66 loc) · 3.19 KB

PageANN

A page-level approximate nearest neighbor search system optimized for SSD-based workloads.

About

PageANN extends Microsoft's DiskANN with page-level graph organization to reduce random I/O and improve SSD utilization during vector search.

Key Differences from DiskANN

  • Page-level indexing: Multiple vectors are merged into disk-aligned pages
  • Reduced I/O: Fewer random accesses during graph traversal
  • SSD-optimized: Better utilization of SSD bandwidth and parallelism

Citation

If you use PageANN in your research, please cite our paper:

Scalable Disk-Based Approximate Nearest Neighbor Search with Page-Aligned Graph

@article{kang2025pageann,
  title={Scalable Disk-Based Approximate Nearest Neighbor Search with Page-Aligned Graph},
  author={Kang, Dingyi and Jiang, Dongming and Yang, Hanshen and Liu, Hang and Li, Bingzhe},
  journal={arXiv preprint arXiv:2509.25487},
  year={2025},
  url={https://www.arxiv.org/abs/2509.25487}
}

Credits

This project is built upon Microsoft Research's DiskANN:

  • DiskANN repository: https://github.com/microsoft/DiskANN
  • Original paper: Subramanya et al., "DiskANN: Fast Accurate Billion-point Nearest Neighbor Search on a Single Node", NeurIPS 2019

Original DiskANN: Copyright (c) Microsoft Corporation PageANN Modifications: Copyright (c) 2025 Dingyi Kang

Author

Dingyi Kang Email: dingyikangosu@gmail.com

License

MIT License - see LICENSE file for details.

This project contains:

  • DiskANN components: Copyright (c) Microsoft Corporation
  • PageANN modifications: Copyright (c) 2025 Dingyi Kang

Building PageANN

Prerequisites

Linux (Ubuntu 20.04+):

sudo apt install make cmake g++ libaio-dev libgoogle-perftools-dev clang-format libboost-all-dev
sudo apt install libmkl-full-dev

Earlier Ubuntu versions: Install Intel MKL manually from oneAPI MKL installer

Build Instructions

mkdir build && cd build
cmake -DCMAKE_BUILD_TYPE=Release ..
make -j

Usage

For detailed workflow and command-line instructions, see:

Key Tools

Main Applications

  • build_vamana_disk_index: Build Vamana vector-level disk index
  • generate_page_graph: Convert Vamana index to page-level graph
  • recommend_vamana_graph_degree: Recommend optimal graph degree parameters
  • search_disk_index: Search the page-level index

Utility Tools

  • compute_groundtruth: Compute ground truth for recall evaluation
  • generate_hash_buckets: Generate hash buckets for routing
  • generate_reorder_pq: Regenerate PQ with different compression levels

Contributing

Contributions are welcome! Please feel free to submit issues or pull requests.

Acknowledgments

Special thanks to Microsoft Research for open-sourcing DiskANN, which forms the foundation of this work.