Accelerating Graph Indexing for ANNS on Modern CPUs

TL;DR

This paper introduces Flash, a novel compact coding strategy that accelerates graph indexing for Approximate Nearest Neighbor Search on modern CPUs by reducing memory access and enhancing SIMD utilization, achieving over 10x speedup.

Contribution

The paper presents Flash, a new compact coding method tailored for graph indexing that significantly improves CPU-based index construction efficiency in high-dimensional ANNS.

Findings

Flash achieves 10.4x to 22.9x speedup in index construction.

It maintains or improves search performance.

It effectively reduces memory access latency and enhances SIMD utilization.

Abstract

In high-dimensional vector spaces, Approximate Nearest Neighbor Search (ANNS) is a key component in database and artificial intelligence infrastructures. Graph-based methods, particularly HNSW, have emerged as leading solutions among various ANNS approaches, offering an impressive trade-off between search efficiency and accuracy. Many modern vector databases utilize graph indexes as their core algorithms, benefiting from various optimizations to enhance search performance. However, the high indexing time associated with graph algorithms poses a significant challenge, especially given the increasing volume of data, query processing complexity, and dynamic index maintenance demand. This has rendered indexing time a critical performance metric for users. In this paper, we comprehensively analyze the underlying causes of the low graph indexing efficiency on modern CPUs, identifying that distance computation dominates indexing time, primarily due to high memory access latency and suboptimal arithmetic operation efficiency. We demonstrate that distance comparisons during index construction can be effectively performed using compact vector codes at an appropriate compression error. Drawing from insights gained through integrating existing compact coding methods in the graph indexing process, we propose a novel compact coding strategy, named Flash, designed explicitly for graph indexing and optimized for modern CPU architectures. By minimizing random memory accesses and maximizing the utilization of SIMD (Single Instruction, Multiple Data) instructions, Flash significantly enhances cache hit rates and arithmetic operations. Extensive experiments conducted on eight real-world datasets, ranging from ten million to one billion vectors, exhibit that Flash achieves a speedup of 10.4$\times$ to 22.9$\times$ in index construction efficiency, while maintaining or improving search performance.