Abstract | ||
---|---|---|
Cluster analysis, such as k-means algorithm, plays a critical role in data mining area, but now it is facing the computational challenge due to the continuously increasing data volume. Parallel computing becomes an efficient way to overcome the difficulty. In this paper, we use Graphics Processing Units (GPU) in the Hadoop framework to accelerate the k-means algorithm. As a result, our algorithm is about 10 times faster than the k-means implemented by Mahout. |
Year | DOI | Venue |
---|---|---|
2014 | 10.1007/978-3-319-11538-2_17 | WEB-AGE INFORMATION MANAGEMENT: WAIM 2014 INTERNATIONAL WORKSHOPS |
Keywords | Field | DocType |
Hadoop, GPU, K-means | Graphics,k-means clustering,Computer science,Parallel computing | Conference |
Volume | ISSN | Citations |
8597 | 0302-9743 | 1 |
PageRank | References | Authors |
0.36 | 7 | 2 |
Name | Order | Citations | PageRank |
---|---|---|---|
Huanxin Zheng | 1 | 1 | 0.36 |
Junmin Wu | 2 | 1 | 0.70 |