Abstract
2014 We present a new approach for combining k-d trees and graphics processing units for nearest neighbor search. It is well known that a direct combination of these tools leads to a non- satisfying performance due to conditional computations and suboptimal memory accesses. To alleviate these problems, we propose a variant of the classical k-d tree data structure, called buffer k-d tree, which can be used to reorganize the search. Our experiments show that we can take advantage of both the hierarchical subdivision induced by k-d trees and the huge computational resources provided by today's many-core devices. We demonstrate the potential of our approach in astronomy, where hundreds of million nearest neighbor queries have to be processed.
Original language | English |
---|---|
Title of host publication | Proceedings of the 31st International Conference on Machine Learning, Beijing, China, 2014 |
Number of pages | 9 |
Publication date | 2014 |
Publication status | Published - 2014 |
Event | International Conference on Machine Learning 2014 - Beijing, China Duration: 21 Jun 2014 → 26 Jun 2014 Conference number: 31 http://icml.cc/2014 |
Conference
Conference | International Conference on Machine Learning 2014 |
---|---|
Number | 31 |
Country/Territory | China |
City | Beijing |
Period | 21/06/2014 → 26/06/2014 |
Internet address |
Series | JMLR: Workshop and Conference Proceedings |
---|---|
Volume | 32 |