Advertisement

Data Mining Algorithms on the Cell Broadband Engine

  • Rubing Duan
  • Alfred Strey
Part of the Lecture Notes in Computer Science book series (LNCS, volume 5168)

Abstract

The Cell Broadband Engine (CBE) is a new heterogeneous multi-core processor from IBM, Sony and Toshiba, and provides the potential to achieve an impressive level of performance for data mining algorithms. In this paper, we describe our implementation of three important classes of data mining algorithms: clustering (k-Means), classification (RBF network), and association rule mining (Apriori) on the CBE. We explain our parallelization methodology and describe the exploitation of thread- and data-level parallelism in each of the three algorithms. Finally we present experimental results on the Cell hardware, where we could achieve a high performance of up to 10 GFLOP/s and a speedup of up to 40.

Keywords

Cell Broadband Engine multi-core k-Means RBF Apriori 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Kahle, J., et al.: Introduction to the Cell multiprocessor. IBM Journal of Research and Development 49(4), 589–604 (2005)CrossRefGoogle Scholar
  2. 2.
    IBM Corporation. Cell Broad Band Engine technology, http://www.alphaworks.ibm.com/topics/cell
  3. 3.
    Bodon, F.: A fast apriori implementation. In: Proceedings of the IEEE ICDM Workshop on Frequent Itemset Mining Implementations (FIMI 2003) (2003)Google Scholar
  4. 4.
    Zaki, M.: Parallel and Distributed Association Mining: A Survey. IEEE Concurrency 7(4), 14–25 (1999)CrossRefGoogle Scholar
  5. 5.
    Williams, S., Shalf, J., Oliker, L., Kamil, S., Husbands, P., Yelick, K.: The potential of the cell processor for scientific computing. In: Proceedings of the 3rd conference on Computing Frontiers (CF 2006), pp. 9–20 (2006)Google Scholar
  6. 6.
    Bader, D., Agarwal, V., Madduriet, K.: On the Design and Analysis of Irregular Algorithms on the Cell Processor: A case study on list ranking. In: 21th IEEE International Parallel and Distributed Processing Symposium (IPDPS) (2007)Google Scholar
  7. 7.
    Buehrer, G., Parthasarathy, S.: The Potential of the Cell Broadband Engine for Data Mining. In: Proceedings of the 33rd Int. Conference on Very Large Data Bases (VLDB) (2007)Google Scholar
  8. 8.
    Bader, D., Agarwal, V.: FFTC: Fastest Fourier Transform for the IBM Cell Broadband Engine. In: Aluru, S., Parashar, M., Badrinath, R., Prasanna, V.K. (eds.) HiPC 2007. LNCS, vol. 4873, pp. 172–184. Springer, Heidelberg (2007)CrossRefGoogle Scholar
  9. 9.
    Petrini, F., et al.: Multicore Surprises: Lessons Learned from Optimizing Sweep3D on the Cell Broadband Engine. In: 21th IEEE International Parallel and Distributed Processing Symposium (IPDPS) (2007)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2008

Authors and Affiliations

  • Rubing Duan
    • 1
  • Alfred Strey
    • 1
  1. 1.Institute of Computer ScienceUniversity of InnsbruckInnsbruckAustria

Personalised recommendations