An Efficient Parallel Algorithm for Self-Organizing Maps using MPI - OpenMP based Cluster
Bhavik Patel, Anurag Jajoo, Yash Tibrewal and Amit Joshi. Article: An Efficient Parallel Algorithm for Self-Organizing Maps using MPI - OpenMP based Cluster. IJCA Proceedings on International Conference on Advanced Computing and Communication Techniques for High Performance Applications ICACCTHPA 2014(2):5-9, February 2015. Full text available. BibTeX
@article{key:article,
author = {Bhavik Patel and Anurag Jajoo and Yash Tibrewal and Amit Joshi},
title = {Article: An Efficient Parallel Algorithm for Self-Organizing Maps using MPI - OpenMP based Cluster},
journal = {IJCA Proceedings on International Conference on Advanced Computing and Communication Techniques for High Performance Applications},
year = {2015},
volume = {ICACCTHPA 2014},
number = {2},
pages = {5-9},
month = {February},
note = {Full text available}
}
Abstract
Cluster Computing is based on the concept that an application can be divided into smaller subtasks which when distributed to different nodes on a cluster (using MPI) will enhance the performance of the application. We can further enhance the performance of that application using a shared programming interface like OpenMP. The Self-Organizing Maps which are extensively used in domains like speech recognition and data classification require considerable amount of time in the training process. This paper proposes a parallel algorithm on a MPI - OpenMP based cluster to reduce the time taken in training and enhance the performance of Self-Organizing Maps (SOM). The results of the algorithm demonstrated a speed-up of 15. 316 as compared to the sequential training of the SOM.
References
- Lawrence, Richard D. , George S. Almasi, and Holly E. Rushmeier. "A scalable parallel algorithm for self-organizing maps with applications to sparse data mining problems. " Data Mining and Knowledge Discovery 3. 2 pp. 171-195, 1999.
- Silva, Bruno, and N. C. Marques. "A hybrid parallel SOM algorithm for large maps in data-mining. " New Trends in Artificial Intelligence (2007).
- Ultsch, Alfred, and H. Peter Siemon. "Kohonen's Self Organizing Feature Maps for Exploratory Data Analysis. " Proc. INNC'90, Int. Neural Network Conf. pp. 305-308, 1990.
- Ultsch, Alfred. "Maps for the visualization of high-dimensional data spaces. " Proc. Workshop on Self organizing Maps. pp. 225-230, 2003.
- Stefanovic, Pavel, and Olga Kurasova. "Visual analysis of self-organizing maps. " Nonlinear Analysis 16. 4 pp. 488-504, 2011.
- Message Passing Interface: MPI http://www. mpi-forum. org?
- MPICH http://www. mpich. org
- OpenMP http://www. openmp. org/?
- UCI Machine Learning Repository http://http://archive. ics. uci. edu/ml/
- Radenski, Atanas. "Shared memory, message passing, and hybrid merge sorts for standalone and clustered smps. " Proc. PDPTA vol. 11, pp. 367-373, 2011.
- Kohonen, Teuvo. "The self-organizing map. " Proceedings of the IEEE 78, no. 9 pp. 1464-1480, 1990.
- Kohonen, Teuvo, Erkki Oja, Olli Simula, Ari Visa, and Jari Kangas. "Engineering applications of the self-organizing map. " Proceedings of the IEEE 84, no. 10 pp. 1358-1384, 1996.