Skip to main content

Revisiting Issues in Benchmark Metric Selection

  • Conference paper
  • First Online:
  • 436 Accesses

Part of the book series: Lecture Notes in Computer Science ((LNPSE,volume 12752))

Abstract

In 1986, Fleming and Wallace presented a case advocating the use of geomean in benchmark results. 23 years later in 2009, Alain Crolotte followed up on that proposal at TPCTC using TPC-D as a reference. Now 11 years later it is time to present another perspective on the age-old argument regarding the best metric for summarizing benchmark data. The aim of this paper is two-fold: (1) summarize the definition and interpretation of the current benchmark metrics for the OLAP family of the TPC benchmarks, including TPC-H, TPC-DS, and TPCx-BB. (2) illustrate the impact and tradeoffs of different statistical measures on the overall benchmark metric score, using both conceptual and data-driven arguments. Our hope is that the paper reinvigorates interest in the benchmark community to re-evaluate the design of benchmark metrics and offer insights that can influence the future direction of benchmark metrics design.

This is a preview of subscription content, log in via an institution.

Buying options

Chapter
USD   29.95
Price excludes VAT (USA)
  • Available as PDF
  • Read on any device
  • Instant download
  • Own it forever
eBook
USD   39.99
Price excludes VAT (USA)
  • Available as EPUB and PDF
  • Read on any device
  • Instant download
  • Own it forever
Softcover Book
USD   54.99
Price excludes VAT (USA)
  • Compact, lightweight edition
  • Dispatched in 3 to 5 business days
  • Free shipping worldwide - see info

Tax calculation will be finalised at checkout

Purchases are for personal use only

Learn about institutional subscriptions

References

  1. ML Benchmark Design Challenges - Hot Chips. https://www.hotchips.org/hc31/HC31_1.9_MethodologyAndMLSystem-MLPerf-rev-b.pdf. Accessed 14 Sept 2020

  2. TPC-H. http://www.tpc.org/tpch/default5.asp. Accessed 14 Sept 2020

  3. TPC-DS. http://www.tpc.org/tpcds/default5.asp. Accessed 14 Sept 2020

  4. TPCx-BB. http://www.tpc.org/tpcx-bb/default5.asp. Accessed 14 Sept 2020

  5. Crolotte, A.: Issues in benchmark metric selection. In: Nambiar, R., Poess, M. (eds.) TPCTC 2009. LNCS, vol. 5895, pp. 146–152. Springer, Heidelberg (2009). https://doi.org/10.1007/978-3-642-10424-4_11

    Chapter  Google Scholar 

  6. Fleming, P.J., Wallace, J.J.: How not to lie with statistics: the correct way to summarize benchmark results. Commun. ACM 29(3), 218–221 (1986)

    Google Scholar 

  7. Vogel, R.M.: The geometric mean?. Commun. Stat. Theor. Methods 1–13 (2020)

    Google Scholar 

  8. Nambiar, R.O., Poess, M.: The Making of TPC-DS. In: VLDB, vol. 6, pp. 1049–1058 (2006)

    Google Scholar 

  9. Mashey, J.R.: War of the benchmark means: time for a truce. ACM SIGARCH Comput. Archit. News 32(4), 1–14 (2004)

    Google Scholar 

  10. John, LK.: More on finding a single number to indicate overall performance of a benchmark suite. ACM SIGARCH Comput. Archit. News 32(1), 3–8 (2004)

    Google Scholar 

  11. Iqbal, M.F., John, L.K.: Confusion by all means. In: Proceedings of the 6th International Workshop on Unique Chips and Systems (UCAS-6). (2010)

    Google Scholar 

  12. Citron, D., Hurani, A., Gnadrey, A.: The harmonic or geometric mean: does it really matter? ACM SIGARCH Comput. Archit. News 34(4), 18–25 (2006)

    Google Scholar 

  13. Three simple statistics for your data visualizations. https://breakforsense.net/three-statistics/. Accessed 14 Sept 2020

  14. TPC-H Google Scholar Search Results. https://scholar.google.com/scholar?as_vis=1&q=tpc-h+&hl=en&as_sdt=1,48. Accessed 14 Sept 2020

  15. TPC-H Results. http://www.tpc.org/tpch/results/tpch_advanced_sort5.asp?PRINTVER=false&FLTCOL1=ALL&ADDFILTERROW=&filterRowCount=1&SRTCOL1=h_sponsor&SRTDIR1=ASC&ADDSORTROW=&sortRowCount=1&DISPRES=100+PERCENT&include_withdrawn_results=none&include_historic_results=yes. Accessed 14 Sept 2020

  16. TPC-H Publication. http://www.tpc.org/tpch/results/tpch_result_detail5.asp?id=119040201. Accessed 14 Sept 2020

  17. TPC-DS Publication. http://www.tpc.org/tpcds/results/tpcds_result_detail5.asp?id=120061701. Accessed 14 Sept 2020

  18. TPCx-BB Publication. http://www.tpc.org/tpcx-bb/results/tpcxbb_result_detail5.asp?id=119101101. Accessed 14 Sept 2020

  19. Cisco UCS Integrated Infrastructure for Big Data. http://www.tpc.org/tpcds/results/tpcds_result_detail5.asp?id=118030501. Accessed 14 Sept 2020

  20. Alibaba Cloud AnalyticDB. http://www.tpc.org/tpcds/results/tpcds_result_detail5.asp?id=120061701. Accessed 14 Sept 2020

  21. TPC. http://www.tpc.org. Accessed 15 Sept 2020

Download references

Author information

Authors and Affiliations

Authors

Corresponding author

Correspondence to Dippy Aggarwal .

Editor information

Editors and Affiliations

Rights and permissions

Reprints and permissions

Copyright information

© 2021 Springer Nature Switzerland AG

About this paper

Check for updates. Verify currency and authenticity via CrossMark

Cite this paper

Elford, C., Aggarwal, D., Shekhar, S. (2021). Revisiting Issues in Benchmark Metric Selection. In: Nambiar, R., Poess, M. (eds) Performance Evaluation and Benchmarking. TPCTC 2020. Lecture Notes in Computer Science(), vol 12752. Springer, Cham. https://doi.org/10.1007/978-3-030-84924-5_3

Download citation

  • DOI: https://doi.org/10.1007/978-3-030-84924-5_3

  • Published:

  • Publisher Name: Springer, Cham

  • Print ISBN: 978-3-030-84923-8

  • Online ISBN: 978-3-030-84924-5

  • eBook Packages: Computer ScienceComputer Science (R0)

Publish with us

Policies and ethics