Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

GPU-Accelerated Algorithms for Graph Vector Search: Taxonomy, Empirical Study, and Research Directions

Yaowen Liu, Xuejia Chen, Anxin Tian, Haoyang Li, Qinbin Li, Xin Zhang, Alexander Zhou, Chen Jason Zhang, Qing Li, Lei Chen

TL;DR

This work presents a comprehensive survey and experimental study of GPU-accelerated graph-based vector search algorithms, establishing a detailed taxonomy of GPU optimization strategies and clarify the mapping between algorithmic tasks and hardware execution units within GPUs.

Abstract

Approximate Nearest Neighbor Search (ANNS) underpins many large-scale data mining and machine learning applications, with efficient retrieval increasingly hinging on GPU acceleration as dataset sizes grow. Although graph-based approaches represent the state of the art in approximate nearest neighbor search, there is a lack of systematic understanding regarding their optimization for modern GPU architectures and their end-to-end effectiveness in practical scenarios. In this work, we present a comprehensive survey and experimental study of GPU-accelerated graph-based vector search algorithms. We establish a detailed taxonomy of GPU optimization strategies and clarify the mapping between algorithmic tasks and hardware execution units within GPUs. Through a thorough evaluation of six leading algorithms on eight large-scale benchmark datasets, we assess both graph index construction and query search performance. Our analysis reveals that distance computation remains the primary computational bottleneck, while data transfer between the host CPU and GPU emerges as the dominant factor influencing real-world latency at large scale. We also highlight key trade-offs in scalability and memory usage across different system designs. Our findings offer clear guidelines for designing scalable and robust GPU-powered approximate nearest neighbor search systems, and provide a comprehensive benchmark for the knowledge discovery and data mining community.

GPU-Accelerated Algorithms for Graph Vector Search: Taxonomy, Empirical Study, and Research Directions

TL;DR

This work presents a comprehensive survey and experimental study of GPU-accelerated graph-based vector search algorithms, establishing a detailed taxonomy of GPU optimization strategies and clarify the mapping between algorithmic tasks and hardware execution units within GPUs.

Abstract

Approximate Nearest Neighbor Search (ANNS) underpins many large-scale data mining and machine learning applications, with efficient retrieval increasingly hinging on GPU acceleration as dataset sizes grow. Although graph-based approaches represent the state of the art in approximate nearest neighbor search, there is a lack of systematic understanding regarding their optimization for modern GPU architectures and their end-to-end effectiveness in practical scenarios. In this work, we present a comprehensive survey and experimental study of GPU-accelerated graph-based vector search algorithms. We establish a detailed taxonomy of GPU optimization strategies and clarify the mapping between algorithmic tasks and hardware execution units within GPUs. Through a thorough evaluation of six leading algorithms on eight large-scale benchmark datasets, we assess both graph index construction and query search performance. Our analysis reveals that distance computation remains the primary computational bottleneck, while data transfer between the host CPU and GPU emerges as the dominant factor influencing real-world latency at large scale. We also highlight key trade-offs in scalability and memory usage across different system designs. Our findings offer clear guidelines for designing scalable and robust GPU-powered approximate nearest neighbor search systems, and provide a comprehensive benchmark for the knowledge discovery and data mining community.
Paper Structure (30 sections, 2 equations, 18 figures, 4 tables, 1 algorithm)

This paper contains 30 sections, 2 equations, 18 figures, 4 tables, 1 algorithm.

Table of Contents

  1. Introduction
  2. Preliminary
  3. Component Analysis
  4. Graph Construction Phase
  5. Search Phase
  6. Experimental Evaluations
  7. Experimental Settings
  8. Graph Construction Evaluation
  9. Construction Efficiency Evaluation
  10. Graph Quality Evaluation
  11. Search Performance
  12. Search with Transfer
  13. Transfer Overhead Analysis
  14. Search without Transfer
  15. Sub-component in Search
  16. ...and 15 more sections

Figures (18)

  • Figure 1: Workflow of GPU-Accelerated ANNS.
  • Figure 2: The pipeline of graph-based ANNS algorithms during the search phase.
  • Figure 3: Average query path length across different algorithms at 90% recall@10.
  • Figure 4: Breakdown of Host-to-Device (HtoD) data transfer overhead (Part 1).
  • Figure 5: The QPS-Recall@10 of GPU-accelerated graph-based ANNS in high-precision regin (top right is better).
  • ...and 13 more figures

Theorems & Definitions (2)

  • definition 1: $k$-Nearest Neighbor Search.
  • definition 2: Approximate $k$-Nearest Neighbor Search