Java as a basis for parallel data mining in workstation clusters

Gimbel M, Philippsen M, Haumacher B, Lockemann PC, Tichy WF (1999)


Publication Language: English

Publication Type: Conference contribution, Original article

Publication year: 1999

Publisher: Springer

Series: Lecture Notes in Computer Science

City/Town: Berlin Heidelberg

Book Volume: 1593

Pages Range: 884-894

Conference Proceedings Title: Proc. 7th International Conference on High Performance Computing and Networking (HPCN Europe)

Event location: Amsterdam, NL NL

ISBN: 978-3-540-65821-4

URI: http://www2.informatik.uni-erlangen.de/publication/download/dataminps.pdf

DOI: 10.1007/BFb0100648

Abstract

The exploitation of hidden information from large datasets by means of data mining techniques suffers from long response times. We address this problem by using the processing power of workstation clusters and have studied the performance of OLAP queries as a first step towards a portable data mining platform.

The results of our study suggest that with the availability of parallel workstation clusters that are equipped with high performance communication networks, fine-grained and communication-intensive parallelizations of queries are promising—even though they are considered too costly in traditional database systems.

The paper describes our Java framework for parallel OLAP-type query execution, necessary optimizations to the standard Java implementation, and analyzes the performance of non-standard parallel execution schemes on a workstation cluster.

Authors with CRIS profile

How to cite

APA:

Gimbel, M., Philippsen, M., Haumacher, B., Lockemann, P.C., & Tichy, W.F. (1999). Java as a basis for parallel data mining in workstation clusters. In Sloot, P. ; Bubak, M. ; Hoekstra, Alfons G. ; Hertzberger, B. (Eds.), Proc. 7th International Conference on High Performance Computing and Networking (HPCN Europe) (pp. 884-894). Amsterdam, NL, NL: Berlin Heidelberg: Springer.

MLA:

Gimbel, Matthias, et al. "Java as a basis for parallel data mining in workstation clusters." Proceedings of the 7th International Conference on High-Performance Computing and Networking (HPCN), Amsterdam, NL Ed. Sloot, P. ; Bubak, M. ; Hoekstra, Alfons G. ; Hertzberger, B., Berlin Heidelberg: Springer, 1999. 884-894.

BibTeX: Download