CRouting: Reducing Expensive Distance Calls in Graph-Based Approximate Nearest Neighbor Search

2025-08-30

AI Summary

To address the high computational cost of distance calculations in graph-based approximate nearest neighbor search (ANNS), this paper proposes CRouting—a novel routing strategy that exploits the concentration-of-measure phenomenon in high-dimensional angular distributions to dynamically skip redundant distance computations during graph traversal. As a lightweight, plug-and-play component, CRouting requires no modification to underlying graph structures and integrates seamlessly with mainstream graph indices (e.g., HNSW, NSG) via minimal code adaptation. Experiments demonstrate that CRouting reduces distance computations by up to 41.5% while preserving recall, yielding up to a 1.48× improvement in query throughput. Its core innovation lies in explicitly modeling high-dimensional geometric properties—specifically, angular distribution—as a principled routing criterion, establishing a new paradigm for computationally efficient graph indexing.

Technology Category

• Planning, Routing, and Scheduling: Routing
• Search and Optimization: Distributed Search
• Machine Learning: Graph-based Machine Learning

Application Category

• Graph Algorithms and Modeling for the Web: Querying, indexing, and retrieval in Web-related graphs
• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics
• Semantics and Knowledge: Scalable techniques for the creation, curation, publication, maintenance, and consumption of large, Web-based, structured, reusable, knowledge graphs and ontologies

Abstract

Approximate nearest neighbor search (ANNS) is a crucial problem in information retrieval and AI applications. Recently, there has been a surge of interest in graph-based ANNS algorithms due to their superior efficiency and accuracy. However, the repeated computation of distances in high-dimensional spaces constitutes the primary time cost of graph-based methods. To accelerate the search, we propose a novel routing strategy named CRouting, which bypasses unnecessary distance computations by exploiting the angle distributions of high-dimensional vectors. CRouting is designed as a plugin to optimize existing graph-based search with minimal code modifications. Our experiments show that CRouting reduces the number of distance computations by up to 41.5% and boosts queries per second by up to 1.48$ imes$ on two predominant graph indexes, HNSW and NSG. Code is publicly available at https://github.com/ISCS-ZJU/CRouting.

Problem

• Reducing expensive distance computations in graph-based ANNS
• Bypassing unnecessary calculations using angle distributions
• Optimizing existing graph search methods with minimal modifications

Innovation

• Bypasses distance computations via angle distributions
• Plugin routing strategy for graph-based search
• Reduces distance calls by up to 41.5%