Crate ruvector_diskann 

ruvector-diskann

DiskANN/Vamana implementation for billion-scale approximate nearest neighbor search.

Algorithm

• Vamana graph: greedy search + α-robust pruning for bounded out-degree
• Product Quantization (PQ): compressed distance for candidate filtering
• Memory-mapped graph: SSD-friendly access, only load neighbors on demand

Reference

Subramanya et al., “DiskANN: Fast Accurate Billion-point Nearest Neighbor Search on a Single Node” (NeurIPS 2019)

Re-exports

pub use error::DiskAnnError;

pub use error::Result;

pub use index::DiskAnnConfig;

pub use index::DiskAnnIndex;

pub use pq::ProductQuantizer;

Modules

distance

Distance computations with SIMD acceleration and optional GPU offload

error

graph

Vamana graph construction with α-robust pruning

index

DiskANN index — ties together Vamana graph, PQ, and mmap persistence

pq

Product Quantization for compressed distance computation