# Theoretically-Efficient and Practical Parallel DBSCAN

arXiv: https://arxiv.org/abs/1912.06255 · HTML: https://mlsanity.com/paper/1912.06255/

## Key facts

- Cited by 8 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 12 Dec 2019.

## Built on by (most cited first, 8 of 8)

- [Fast Parallel Algorithms for Euclidean Minimum Spanning Tree and Hierarchical Spatial Clustering](https://mlsanity.com/paper/2104.01126/) (2021)
- [PECANN: Parallel Efficient Clustering with Graph-Based Approximate Nearest Neighbor Search](https://mlsanity.com/paper/2312.03940/) (2023)
- [Enhancing Maritime Trajectory Forecasting via H3 Index and Causal Language Modelling (CLM)](https://mlsanity.com/paper/2405.09596/) (2024)
- [Adaptive and Robust DBSCAN with Multi-agent Reinforcement Learning](https://mlsanity.com/paper/2505.04339/) (2025)
- [Learned k-NN Distance Estimation](https://mlsanity.com/paper/2208.14210/) (2022)
- [Distributed Clustering based on Distributional Kernel](https://mlsanity.com/paper/2409.09418/) (2024)
- [SACA: Selective Attention-Based Clustering Algorithm](https://mlsanity.com/paper/2508.17150/) (2025)
- [LiGuard: A Streamlined Open-Source Framework for Rapid & Interactive Lidar Research](https://mlsanity.com/paper/2509.02902/) (2025)

## Abstract (arXiv)

The DBSCAN method for spatial clustering has received significant attention due to its applicability in a variety of data analysis tasks. There are fast sequential algorithms for DBSCAN in Euclidean space that take O(nlog n) work for two dimensions, sub-quadratic work for three or more dimensions, and can be computed approximately in linear work for any constant number of dimensions. However, existing parallel DBSCAN algorithms require quadratic work in the worst case, making them inefficient for large datasets. This paper bridges the gap between theory and practice of parallel DBSCAN by presenting new parallel algorithms for Euclidean exact DBSCAN and approximate DBSCAN that match the work bounds of their sequential counterparts, and are highly parallel (polylogarithmic depth). We present implementations of our algorithms along with optimizations that improve their practical performance. We perform a comprehensive experimental evaluation of our algorithms on a variety of datasets and parameter settings. Our experiments on a 36-core machine with hyper-threading show that we outperform existing parallel DBSCAN implementations by up to several orders of magnitude, and achieve speedups by up to 33x over the best sequential algorithms.