RT-HiSS: Ray Tracing Accelerated High Dimensional Vector Similarity Searches

2026-09-01

AI Summary

本文提出RT-HiSS算法，利用GPU的光线追踪核心加速高维数据集的相似性搜索，通过快速索引构建、候选优化等方法提高搜索效率。

Abstract

Recent GPU generations include special-purpose ray tracing (RT) cores for graphics applications. While RT cores are primarily used for rendering, recent works show they can be leveraged for general-purpose tasks, including similarity searches. However, existing approaches do not support datasets exceeding three dimensions. In this work, we propose RT-HiSS, the first exact GPU RT-core-based similarity search algorithm for high-dimensional datasets. GPU similarity search often scales poorly for large datasets with substantial search distances. To address this, RT-HiSS uses RT cores for fast index construction and searches, followed by candidate refinement on CUDA cores. We introduce a two-pass approach to estimate an upper bound on result size, enabling efficient batching under GPU memory constraints with near-perfect load balancing. Additionally, we examine shared memory tiling and compressed result masks to improve GPU resource utilization. RT-HiSS yields speedups up to 8.37$\times$ over competitive state-of-the-art GPU algorithms and up to 2,368.26$\times$ relative to the brute-force algorithm across six real-world datasets.

Problem

• high-dimensional datasets
• similarity search
• GPU
• ray tracing cores

Innovation

• RT cores
• high-dimensional datasets
• similarity search
• CUDA cores
• shared memory tiling