# JUNO: Optimizing High-Dimensional Approximate Nearest Neighbour Search with Sparsity-Aware Algorithm and Ray-Tracing Core Mapping

arXiv: https://arxiv.org/abs/2312.01712 · HTML: https://mlsanity.com/paper/2312.01712/

## Key facts

- Cited by 5 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 4 Dec 2023.

## Built on by (most cited first, 5 of 5)

- [RAGO: Systematic Performance Optimization for Retrieval-Augmented Generation Serving](https://mlsanity.com/paper/2503.14649/) (2025)
- [TeleRAG: Efficient Retrieval-Augmented Generation Inference with Lookahead Retrieval](https://mlsanity.com/paper/2502.20969/) (2025)
- [ClusterFusion: Expanding Operator Fusion Scope for LLM Inference via Cluster-Level Collective Primitive](https://mlsanity.com/paper/2508.18850/) (2025)
- [UpANNS: Enhancing Billion-Scale ANNS Efficiency with Real-World PIM Architecture](https://mlsanity.com/paper/2410.23805/) (2024)
- [Scalable High-Fidelity Macromolecular Docking for GPU-Accelerated Supercomputers](https://mlsanity.com/paper/2608.07078/) (2026)

## Abstract (arXiv)

Approximate nearest neighbor (ANN) search is a widely applied technique in modern intelligent applications, such as recommendation systems and vector databases. Therefore, efficient and high-throughput execution of ANN search has become increasingly important. In this paper, we first characterize the state-of-the-art product quantization-based method of ANN search and identify a significant source of inefficiency in the form of unnecessary pairwise distance calculations and accumulations. To improve efficiency, we propose JUNO, an end-to-end ANN search system that adopts a carefully designed sparsity- and locality-aware search algorithm. We also present an efficient hardware mapping that utilizes ray tracing cores in modern GPUs with pipelined execution on tensor cores to execute our sparsity-aware ANN search algorithm. Our evaluations on four datasets ranging in size from 1 to 100 million search points demonstrate 2.2x-8.5x improvements in search throughput. Moreover, our algorithmic enhancements alone achieve a maximal 2.6x improvement on the hardware without the acceleration of the RT core.