GRNND: A GPU-Parallel Relative NN-Descent Algorithm for Efficient Approximate Nearest Neighbor Graph Construction

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Li, Xiang, Chang, Qiong, Li, Yun, Miyazaki, Jun
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866912625920573440
author Li, Xiang
Chang, Qiong
Li, Yun
Miyazaki, Jun
author_facet Li, Xiang
Chang, Qiong
Li, Yun
Miyazaki, Jun
contents Relative Nearest Neighbor Descent (RNN-Descent) is a state-of-the-art algorithm for constructing sparse approximate nearest neighbor (ANN) graphs by combining the iterative refinement of NN-Descent with the edge-pruning rules of the Relative Neighborhood Graph (RNG). It has demonstrated strong effectiveness in large-scale search tasks such as information retrieval and related tasks. However, as the amount and dimensionality of data increase, the complexity of graph construction in RNN-Descent rises sharply, making this stage increasingly time-consuming and even prohibitive for subsequent query processing. In this paper, we propose GRNND, the first GPU-parallel algorithm of RNN-Descent designed to fully exploit GPU architecture. GRNND introduces a disordered neighbor propagation strategy to mitigate synchronized update traps, enhancing structural diversity, and avoiding premature convergence during parallel execution. It also leverages warp-level cooperative operations and a double-buffered neighbor pool with fixed capacity for efficient memory access, eliminate contention, and enable highly parallelized neighbor updates. Extensive experiments demonstrate that GRNND consistently outperforms existing CPU- and GPU-based methods. GRNND achieves 2.4 to 51.7x speedup over existing GPU methods, and 17.8 to 49.8x speedup over CPU methods.
format Preprint
id arxiv_https___arxiv_org_abs_2510_02774
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle GRNND: A GPU-Parallel Relative NN-Descent Algorithm for Efficient Approximate Nearest Neighbor Graph Construction
Li, Xiang
Chang, Qiong
Li, Yun
Miyazaki, Jun
Distributed, Parallel, and Cluster Computing
Relative Nearest Neighbor Descent (RNN-Descent) is a state-of-the-art algorithm for constructing sparse approximate nearest neighbor (ANN) graphs by combining the iterative refinement of NN-Descent with the edge-pruning rules of the Relative Neighborhood Graph (RNG). It has demonstrated strong effectiveness in large-scale search tasks such as information retrieval and related tasks. However, as the amount and dimensionality of data increase, the complexity of graph construction in RNN-Descent rises sharply, making this stage increasingly time-consuming and even prohibitive for subsequent query processing. In this paper, we propose GRNND, the first GPU-parallel algorithm of RNN-Descent designed to fully exploit GPU architecture. GRNND introduces a disordered neighbor propagation strategy to mitigate synchronized update traps, enhancing structural diversity, and avoiding premature convergence during parallel execution. It also leverages warp-level cooperative operations and a double-buffered neighbor pool with fixed capacity for efficient memory access, eliminate contention, and enable highly parallelized neighbor updates. Extensive experiments demonstrate that GRNND consistently outperforms existing CPU- and GPU-based methods. GRNND achieves 2.4 to 51.7x speedup over existing GPU methods, and 17.8 to 49.8x speedup over CPU methods.
title GRNND: A GPU-Parallel Relative NN-Descent Algorithm for Efficient Approximate Nearest Neighbor Graph Construction
topic Distributed, Parallel, and Cluster Computing
url https://arxiv.org/abs/2510.02774