Lossless Compression of Proximity Graph Indexes for Efficient Approximate Nearest Neighbors Search
This paper presents a compression technique for HNSW graph-based ANN indexes that achieves significant memory savings with negligible query latency.
This paper presents a novel compression technique for HNSW graph-based ANN indexes that achieves significant memory savings with negligible query latency, improving upon prior work that may have compromised query latency for the sake of compression.
Before reading this…
Applications
- →Vector search
- →Graph-based ANN indexes
To understand this paper, make sure you know these concepts first:
- Knowledge of graph-based ANN indexesfind papers →
- Understanding of compression techniquesfind papers →
Abstract
More Like ThisGraph-based Approximate Nearest Neighbor (ANN) indexes, such as HNSW, are the backbone of modern vector search. However, their massive memory footprint creates a severe infrastructure bottleneck. While vector quantization is widely adopted to reduce the memory required to store data vectors, the graph's edge structure increasingly dominates total memory usage. Compressing these adjacency lists for online search presents a critical system challenge: because decompression occurs during query processing, i.e., while navigating the graph, memory savings cannot come at the expense of query latency. In this paper, we present a compression technique explicitly tailored for live HNSW routing. By applying offline graph permutations to artificially induce structural locality, we enable the use of SIMD-accelerated integer compressors, such as Stream VByte, for rapid on-the-fly decompression. Evaluated across diverse real-world workloads, we significantly outperform the current state of the art in both compression ratio and query latency. Our experimental results show that we achieve compression ratios up to 52.7% with negligible penalties in latency compared to the uncompressed baseline. Furthermore, our technique has been effectively deployed in production at eBay, achieving up to 44.3% reduction in edge storage with no degradation in query latency.