This title appears in the Scientific Report :
2015
Please use the identifier:
http://dx.doi.org/10.1145/2834892.2834894 in citations.
HPDBSCAN - Highly parallel DBSCAN
HPDBSCAN - Highly parallel DBSCAN
Clustering algorithms in the field of data-mining are used to aggregate similar objects into common groups. One of the best-known of these algorithms is called DBSCAN. Its distinct design enables the search for an apriori unknown number of arbitrarily shaped clusters, and at the same time allows to...
Saved in:
Personal Name(s): | Götz, Markus (Corresponding author) |
---|---|
Bodenstein, Christian / Riedel, Morris | |
Contributing Institute: |
Jülich Supercomputing Center; JSC |
Published in: | S. 2 |
Published in: |
Proceedings of the Workshop on Machine Learning in High-Performance Computing Environments - MLHPC '15 |
Imprint: |
ACM Press New York, New York, USA
2015
|
Physical Description: |
10p. |
ISBN: |
978-1-4503-4006-9 |
DOI: |
10.1145/2834892.2834894 |
Conference: | Workshop Workshop on Machine Learning in High-Performance Computing Environments, subworkshop to Supercomputing 2015, Austin (Texas), 2015-11-15 - 2015-11-15 |
Document Type: |
Contribution to a conference proceedings |
Research Program: |
Data-Intensive Science and Federated Computing |
Publikationsportal JuSER |
Clustering algorithms in the field of data-mining are used to aggregate similar objects into common groups. One of the best-known of these algorithms is called DBSCAN. Its distinct design enables the search for an apriori unknown number of arbitrarily shaped clusters, and at the same time allows to filter out noise. Due to its sequential formulation, the parallelization of DBSCAN renders a challenge. In this paper we present a new parallel approach which we call HPDBSCAN. It employs three major techniques in order to break the sequentiality, empower workload-balancing as well as speed up neighborhood searches in distributed parallel processing environments i) a computation split heuristic for domain decomposition, ii) a data index preprocessing step and iii) a rule-based cluster merging scheme.As a proof-of-concept we implemented HPDBSCAN as an OpenMP/MPI hybrid application. Using real-world data sets, such as a point cloud from the old town of Bremen, Germany, we demonstrate that our implementation is able to achieve a significant speed-up and scale-up in common HPC setups. Moreover, we compare our approach with previous attempts to parallelize DBSCAN showing an order of magnitude improvement in terms of computation time and memory consumption. |