Advanced search
1 file | 455.68 KB Add to list

Ranking commercial machines through data transposition

Author
Organization
Abstract
The performance numbers reported by benchmarking consortia and corporations provide little or no insight into the performance of applications of interest that are not part of the benchmark suite. This paper describes data transposition, a novel methodology for addressing this ubiquitous benchmarking problem. Data transposition predicts the performance for an application of interest on a target machine based on its performance similarities with the industry-standard benchmarks on a limited number of predictive machines. The key idea of data transposition is to exploit machine similarity rather than workload similarity as done in prior work, i.e., data transposition identifies a predictive machine that is most similar to the target machine of interest for predicting performance for the application of interest. We demonstrate the accuracy and effectiveness of data transposition using the SPEC CPU2006 benchmarks and a set of 117 commercial machines. We report that the machine ranking obtained through data transposition correlates well with the machine ranking obtained using measured performance numbers (average correlation coefficient of 0.93). Not only does data transposition improve average correlation, we also demonstrate that data transposition is more robust towards outlier benchmarks, i.e., the worst-case correlation coefficient improves from 0.59 by prior art to 0.71. More concretely, using data transposition to predict the top-1 machine for an application of interest leads to the best performing machine for most workloads (average deficiency of 1.2% and max deficiency of 24.8% for one benchmark), whereas prior work leads to deficiencies over 100% for some workloads.
Keywords
computer architecture, performance prediction, workload characterization

Downloads

  • iiswc11 2.pdf
    • full text
    • |
    • open access
    • |
    • PDF
    • |
    • 455.68 KB

Citation

Please use this url to cite or link to this publication:

MLA
Piccart, Beau et al. “Ranking Commercial Machines Through Data Transposition.” International Symposium on Workload Characterization Proceedings. New York, NY, USA: IEEE, 2011. 3–14. Print.
APA
Piccart, B., Georges, A., Blockeel, H., & Eeckhout, L. (2011). Ranking commercial machines through data transposition. International Symposium on Workload Characterization Proceedings (pp. 3–14). Presented at the 2011 IEEE International symposium on Workload Characterization (IISWC 2011), New York, NY, USA: IEEE.
Chicago author-date
Piccart, Beau, Andy Georges, Hendrik Blockeel, and Lieven Eeckhout. 2011. “Ranking Commercial Machines Through Data Transposition.” In International Symposium on Workload Characterization Proceedings, 3–14. New York, NY, USA: IEEE.
Chicago author-date (all authors)
Piccart, Beau, Andy Georges, Hendrik Blockeel, and Lieven Eeckhout. 2011. “Ranking Commercial Machines Through Data Transposition.” In International Symposium on Workload Characterization Proceedings, 3–14. New York, NY, USA: IEEE.
Vancouver
1.
Piccart B, Georges A, Blockeel H, Eeckhout L. Ranking commercial machines through data transposition. International Symposium on Workload Characterization Proceedings. New York, NY, USA: IEEE; 2011. p. 3–14.
IEEE
[1]
B. Piccart, A. Georges, H. Blockeel, and L. Eeckhout, “Ranking commercial machines through data transposition,” in International Symposium on Workload Characterization Proceedings, Austi, TX, USA, 2011, pp. 3–14.
@inproceedings{1977509,
  abstract     = {The performance numbers reported by benchmarking consortia and corporations provide little or no insight into the performance of applications of interest that are not part of the benchmark suite. This paper describes data transposition, a novel methodology for addressing this ubiquitous benchmarking problem. Data transposition predicts the performance for an application of interest on a target machine based on its performance similarities with the industry-standard benchmarks on a limited number of predictive machines. The key idea of data transposition is to exploit machine similarity rather than workload similarity as done in prior work, i.e., data transposition identifies a predictive machine that is most similar to the target machine of interest for predicting performance for the application of interest.
We demonstrate the accuracy and effectiveness of data transposition using the SPEC CPU2006 benchmarks and a set of 117 commercial machines. We report that the machine ranking obtained through data transposition correlates well with the machine ranking obtained using measured performance numbers (average correlation coefficient of 0.93). Not only does data transposition improve average correlation, we also demonstrate that data transposition is more robust towards outlier benchmarks, i.e., the worst-case correlation coefficient improves from 0.59 by prior art to 0.71. More concretely, using data transposition to predict the top-1 machine for an application of interest leads to the best performing machine for most workloads (average deficiency of 1.2% and max deficiency of 24.8% for one benchmark), whereas prior work leads to deficiencies over 100% for some workloads.},
  author       = {Piccart, Beau and Georges, Andy and Blockeel, Hendrik and Eeckhout, Lieven},
  booktitle    = {International Symposium on Workload Characterization Proceedings},
  isbn         = {9781457720642},
  keywords     = {computer architecture,performance prediction,workload characterization},
  language     = {eng},
  location     = {Austi, TX, USA},
  pages        = {3--14},
  publisher    = {IEEE},
  title        = {Ranking commercial machines through data transposition},
  url          = {http://dx.doi.org/10.1109/IISWC.2011.6114192},
  year         = {2011},
}

Altmetric
View in Altmetric
Web of Science
Times cited: