VSAG: An Optimized Search Framework for Graph-based Approximate Nearest Neighbor Search

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Zhong, Xiaoyao, Li, Haotian, Jin, Jiabao, Yang, Mingyu, Chu, Deming, Wang, Xiangyu, Shen, Zhitao, Jia, Wei, Gu, George, Xie, Yi, Lin, Xuemin, Shen, Heng Tao, Song, Jingkuan, Cheng, Peng
Format:	Preprint
Published:	2025
Subjects:	Databases
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

_version_	1866911056995024896
author	Zhong, Xiaoyao Li, Haotian Jin, Jiabao Yang, Mingyu Chu, Deming Wang, Xiangyu Shen, Zhitao Jia, Wei Gu, George Xie, Yi Lin, Xuemin Shen, Heng Tao Song, Jingkuan Cheng, Peng
author_facet	Zhong, Xiaoyao Li, Haotian Jin, Jiabao Yang, Mingyu Chu, Deming Wang, Xiangyu Shen, Zhitao Jia, Wei Gu, George Xie, Yi Lin, Xuemin Shen, Heng Tao Song, Jingkuan Cheng, Peng
contents	Approximate nearest neighbor search (ANNS) is a fundamental problem in vector databases and AI infrastructures. Recent graph-based ANNS algorithms have achieved high search accuracy with practical efficiency. Despite the advancements, these algorithms still face performance bottlenecks in production, due to the random memory access patterns of graph-based search and the high computational overheads of vector distance. In addition, the performance of a graph-based ANNS algorithm is highly sensitive to parameters, while selecting the optimal parameters is cost-prohibitive, e.g., manual tuning requires repeatedly re-building the index. This paper introduces VSAG, an open-source framework that aims to enhance the in production performance of graph-based ANNS algorithms. VSAG has been deployed at scale in the services of Ant Group, and it incorporates three key optimizations: (i) efficient memory access: it reduces L3 cache misses with pre-fetching and cache-friendly vector organization; (ii) automated parameter tuning: it automatically selects performance-optimal parameters without requiring index rebuilding; (iii) efficient distance computation: it leverages modern hardware, scalar quantization, and smartly switches to low-precision representation to dramatically reduce the distance computation costs. We evaluate VSAG on real-world datasets. The experimental results show that VSAG achieves the state-of-the-art performance and provides up to 4x speedup over HNSWlib (an industry-standard library) while ensuring the same accuracy.
format	Preprint
id	arxiv_https___arxiv_org_abs_2503_17911
institution	arXiv
publishDate	2025
record_format	arxiv
spellingShingle	VSAG: An Optimized Search Framework for Graph-based Approximate Nearest Neighbor Search Zhong, Xiaoyao Li, Haotian Jin, Jiabao Yang, Mingyu Chu, Deming Wang, Xiangyu Shen, Zhitao Jia, Wei Gu, George Xie, Yi Lin, Xuemin Shen, Heng Tao Song, Jingkuan Cheng, Peng Databases Approximate nearest neighbor search (ANNS) is a fundamental problem in vector databases and AI infrastructures. Recent graph-based ANNS algorithms have achieved high search accuracy with practical efficiency. Despite the advancements, these algorithms still face performance bottlenecks in production, due to the random memory access patterns of graph-based search and the high computational overheads of vector distance. In addition, the performance of a graph-based ANNS algorithm is highly sensitive to parameters, while selecting the optimal parameters is cost-prohibitive, e.g., manual tuning requires repeatedly re-building the index. This paper introduces VSAG, an open-source framework that aims to enhance the in production performance of graph-based ANNS algorithms. VSAG has been deployed at scale in the services of Ant Group, and it incorporates three key optimizations: (i) efficient memory access: it reduces L3 cache misses with pre-fetching and cache-friendly vector organization; (ii) automated parameter tuning: it automatically selects performance-optimal parameters without requiring index rebuilding; (iii) efficient distance computation: it leverages modern hardware, scalar quantization, and smartly switches to low-precision representation to dramatically reduce the distance computation costs. We evaluate VSAG on real-world datasets. The experimental results show that VSAG achieves the state-of-the-art performance and provides up to 4x speedup over HNSWlib (an industry-standard library) while ensuring the same accuracy.
title	VSAG: An Optimized Search Framework for Graph-based Approximate Nearest Neighbor Search
topic	Databases
url	https://arxiv.org/abs/2503.17911