Browse > Article
http://dx.doi.org/10.3837/tiis.2019.08.018

Composite Measures of Supercomputer Technology  

Kim, Nam-Gyu (Division of National Supercomputing, Korea Institute of Science and Technology Information)
On, Noo Ri (Division of National Supercomputing, Korea Institute of Science and Technology Information)
Koh, Myoung-Ju (Division of National Supercomputing, Korea Institute of Science and Technology Information)
Lee, JongSuk Ruth (Division of National Supercomputing, Korea Institute of Science and Technology Information)
Cho, Keun-Tae (Department of Management of Technology, Sungkyunkwan University)
Publication Information
KSII Transactions on Internet and Information Systems (TIIS) / v.13, no.8, 2019 , pp. 4142-4159 More about this Journal
Abstract
We have developed composite measures of supercomputer technology, reflecting various factors of supercomputers using Martino's scoring model. CPUs, accelerators, memory, interconnection networks, and power consumption are chosen as factors of the model. The weight values of the factors are derived based on a survey of 129 domestic and international experts. The measured values are then standardized to integrate measurement units of the factors in the model. This model has been applied to 50 supercomputers, and rank correlation analysis was performed using representative measures. As a consequence, the ranking drastically changes except for the 1st and 2nd supercomputers on the TOP500. In addition, the characteristics of memory and interconnection networks influence the ranking, and the results demonstrate that the proposed model has low correlations with HPL and HPCG but a high correlation with Green500. This indicates that power consumption is an important factor that has a significant effect on the measures of supercomputer technology. In addition, it is determined that the differences between the HPL ranking and the proposed model ranking are influenced by power consumption, CPU theoretical peak performance, and main memory bandwidth in order of significance. In conclusion, the composite measures proposed in this study are more suitable for comprehensively describing supercomputer technology than existing performance measures. The findings of this study are expected to support decision making related to management and policy in the procurement and operation of supercomputers.
Keywords
Composite Measures; Measures of Technology; Scoring Model; Supercomputer Performance; Supercomputer policy;
Citations & Related Records
연도 인용수 순위
  • Reference
1 Gene H. Golub and James M. Ortega, Scientific Computing and Differential Equations: An Introduction to Numerical Methods, 1st Edition, Academic Press, Orlando, 1991.
2 E. Gallopoulos and A. Sameh, "CSE: content and product," IEEE Computational Science and Engineering, vol. 4, no. 2, pp. 39-43, June, 1997.   DOI
3 PITAC (President's Information Technology Advisory Committee), Computational Science: Ensuring America's Competitiveness, the National Coordination Office for Information Technology Research and Development, Arlington, 2005.
4 National Research Council, Getting Up to Speed: The Future of Supercomputing, The National Academies Press, Washington, DC, 2005.
5 Jack J. Dongarra, Piotr Luszczek and Antoine Petitet, "The LINPACK benchmark: past, present and future," Concurrency and Computation: Practice and Experience, vol. 15, no. 9, pp. 803-820, August, 2003.   DOI
6 The TOP500 Supercomputer Sites. https://www.top500.org
7 National Academies of Sciences, Engineering, and Medicine, Future Directions for NSF Advanced Computing Infrastructure to Support U.S. Science and Engineering in 2017-2020, The National Academies Press, Washington, DC, 2016.
8 Xiang Chen, Meizhuo Zhang, Minghui Wang, Wensheng Zhu, Kelly Cho and Heping Zhang, "Memory management in genome-wide association studies," BMC Proceedings, vol. 3, no. 7, pp. S54, December, 2009.   DOI
9 Tzu-Yi Chen, Omid Khalili, Roy L. Campbell, Laura Carrington, Mustafa M. Tikir and Allan Snavely, "Performance Prediction and Ranking of Supercomputers," Advances in Computers, Vol. 72, pp. 135-172, 2008.   DOI
10 John L. Gustafson and Rajat Todi, "Conventional Benchmarks as a Sample of the Performance Spectrum," Journal of Supercomputing, pp 321-342, vol. 13, no. 3, May 1999.   DOI
11 Ryusuke Egawa, Kazuhiko Komatsu, Yoko Isobe, Toshihiro Kato, Souya Fujimoto, Hiroyuki Takizawa, Akihiro Musa and Hiroaki Kobayashi, "Performance and Power Analysis of SX-ACE using HP-X Benchmark Programs," in Proc. of IEEE Int. Conf. on Cluster Computing, pp. 693-700, September 5-8, 2017.
12 Vladimir Marjanovic, Jose Gracia and Colin W. Glass, "HPC Benchmarking: Problem Size Matters," in Proc. of Int. Workshop on Performance Modeling, Benchmarking and Simulation of High Performance Computing Systems, pp. 1-10, November 14, 2016.
13 Wu-chun Feng, Xizhou Feng and Rong Ge, "Green Supercomputing Comes of Age," IT Professional, vol. 10, no. 1, pp. 17-23, January, 2008.   DOI
14 Jeffrey S. Vetter, Contemporary High Performance Computing: From Petascale toward Exascale, 1st Edition, Chapman and Hall, New York, 2013.
15 Francesco Fraternali, Andrea Bartolini, Carlo Cavazzoni and Luca Benini, "Quantifying the Impact of Variability and Heterogeneity on the Energy Efficiency for a Next-Generation Ultra-Green Supercomputer," IEEE Transactions on Parallel and Distributed Systems, vol. 29, no. 7, pp. 1575-1588, July, 2018.   DOI
16 Stephen W. Keckler, William J. Dally, Brucek Khailany, Michael Garland and David Glasco, "GPUs and the Future of Parallel Computing," IEEE Micro, vol. 31, no. 5, pp. 7-17, September, 2011.   DOI
17 J.R. Neely, "The U.S. Government Role in HPC: Technology, Mission, and Policy," in Proc. of Conf. on Comparing High Performance Computing in the U.S. and China, pp.1-16, April 29-30, 2014.
18 David Rohr, Sebastian Kalcher, Matthias Bach, Abdulqadir A. Alaqeeliy, Hani M. Alzaidy, Dominic Eschweiler, Volker Lindenstruth, Sakhar B. Alkhereyfy, Ahmad Alharthiy, Abdulelah Almubaraky, Ibraheem Alqwaizy and Riman Bin Suliman, "An Energy-Efficient Multi-GPU Supercomputer," in Proc. of IEEE Int. Conf. on High Performance Computing and Communications, pp.42-45, August 20-22, 2014.
19 David Rohr, Gvozden Neskovic and Volker Lindenstruth, "The L-CSC cluster: Optimizing power efficiency to become the greenest supercomputer in the world in the Green500 list of November 2014," The Journal of Supercomputing Frontiers and Innovations, vol. 2, no. 3, pp.41-48, July, 2015.
20 Tapasya Patki, David K. Lowenthal, Barry Rountree, Martin Schulz, and Bronis R. de Supinski, "Exploring hardware overprovisioning in power-constrained, high performance computing," in Proc. of Int. ACM Conf. on supercomputing, pp.173-182, June 10-14, 2013.
21 The Green500. https://www.top500.org/green500
22 Everett Phillips and Massimiliano Fatica, "A CUDA Implementation of the High Performance Conjugate Gradient Benchmark," in Proc. of High Performance Computing Systems. Performance Modeling, Benchmarking, and Simulation, Springer International Publishing, pp. 68-84, 2015.
23 Jack Dongarra, Michael A Heroux and Piotr Luszczek, "High-performance conjugate-gradient benchmark: A new metric for ranking high-performance computing systems," The International Journal of High Performance Computing Applications, vol. 30, no. 1, pp. 3-10, February, 2016.   DOI
24 Shinobu Yoshimura, Muneo Hori and Makoto Ohsaki, High-Performance Computing for Structural Mechanics and Earthquake/Tsunami Engineering, 1st Edition, Springer International Publishing, Switzerland, 2016.
25 JaeHyuk Kwack and Gregory H. Bauer, "HPCG and HPGMG benchmark tests on multiple program, multiple data (MPMD) mode on Blue Waters-A Cray XE6/XK7 hybrid system," Concurrency and Computation: Practice and Experience, vol. 30, no. 1, October, 2017.
26 Joseph P. Martino, "A comparison of two composite measures of technology," Technological Forecasting and Social Change, vol. 44, no. 2, pp. 147-159, September, 1993.   DOI
27 Joseph P. Martino, Technological Forecasting for Decision Making, 3rd Edition, McGraw-Hill, New York, 1993.
28 Thomas Sterling, Matthew Anderson and Maciej Brodowicz, High Performance Computing - Modern Systems and Practices, 1st Edition, Morgan Kaufmann, San Francisco, 2017.
29 Sun Yeong Heo, Duk Joon Chang, Jae Kyoung Shin, "Ordering Items from Ranking Procedures in Survey Research," Survey Research, vol. 9, no. 2, pp.29-49, July, 2008.
30 Solmaz Salehian and Yonghong Yan, "Evaluation of Knight Landing High Bandwidth Memory for HPC Workloads," in Proc. of Workshop on Irregular Applications: Architectures and Algorithms, pp.10:1-10:4, November 12-17, 2017.
31 FUJITSU Supercomputer PRIMEHPC FX100. http://www.fujitsu.com/global/products/computing/servers/supercomputer/primehpc-fx100/
32 Yuichiro Ajima, Takahiro Kawashima, Takayuki Okamoto, Naoyuki Shida, Kouichi Hirai, Toshiyuki Shimizu, Shinya Hiramoto, Yoshiro Ikeda, Takahide Yoshikawa, Kenji Uchida and Tomohiro Inoue, "The Tofu Interconnect D," in Proc. of IEEE Int. Conf. on Cluster Computing, pp.646-654, September 10-13, 2018.
33 Kim Soon-Kwi, Logistic regression analysis, Kyowoo, Seoul, 2014.
34 Supercomputer at the Information Technology Center of Nagoya University. http://www.icts.nagoya-u.ac.jp/en/sc/
35 Alfred P. Rovai, Jason D. Baker and Michael K. Ponton, Social Science Research Design and Statistics: A Practitioner's Guide to Research Methods and IBM SPSS, 2nd Edition, Watertree Press, Chesapeake, 2013.