ABSTRACT
Efficiency in supercomputing has traditionally focused on execution time. In early 2000's, the concept of total cost of ownership was re-introduced, with the introduction of efficiency measure to include aspects such as energy and space. Yet the supercomputing community has never agreed upon a metric that can cover these aspects completely and also provide a fair basis for comparison. This paper examines the metrics that have been proposed in the past decade, and proposes a vector-valued metric for efficient supercomputing. Using this metric, the paper presents a study of where the supercomputing industry has been and where it stands today with respect to efficient supercomputing.
- L. Barroso and U. Hölzle. The case for energy-proportional computing. IEEE Computer, 40(12):33--37, Dec. 2007. Google ScholarDigital Library
- C. Bekas and A. Curioni. A new energy aware performance metric. In International Conference on Energy-Aware High Performance Computing, Sept. 2010.Google ScholarCross Ref
- K. D. Bois, T. Schaeps, S. Polfliet, F. Ryckbosch, and L. Eeckhout. SWEEP: Evaluating computer system energy efficiency using synthetic workloads. In International Conference on High Performance Embedded Architectures and Compilers, Jan. 2011. Google ScholarDigital Library
- D. Brooks, P. Bose, S. Schuster, H. Jacobson, P. Kudva, A. Buyuktosunoglu, J.-D. Wellman, V. Zyuban, M. Gupta, and P. Cook. Power-aware microarchitecture: Design and modeling challenges for the next generation microprocessors. IEEE Micro, 20(6):26--44, November/December 2000. Google ScholarDigital Library
- T. Conte and W. Hwu. Advances in benchmarking techniques: New standards and quantitative metrics. Advances in Computers, 41:231--253, 1995.Google ScholarCross Ref
- J. Dongarra, P. Luszczek, and A. Petitet. The LINPACK benchmark: Past, present, and future. Concurrency and Computation: Practice and Experience, 15(9):803--820, Aug. 2003.Google ScholarCross Ref
- Energy Efficient High Performance Computing Working Group. http://eehpcwg.lbl.gov.Google Scholar
- A. Fanara, E. Haines, and A. Howard. The state of energy and performance benchmarking for enterprise servers. In TPC Technology Conference on Performance Evaluation and Benchmarking, Aug. 2009. Google ScholarDigital Library
- R. Ge, X. Feng, and K. Cameron. Improvement of power-performance efficiency for high-end computing. In Workshop on High-Performance, Power-Aware Computing, Apr. 2005. Google ScholarDigital Library
- R. Gonzalez and M. Horowitz. Energy dissipation in general purpose microprocessors. IEEE Journal of Solid-State Circuits, 31(9):1277--1284, Sept. 1996.Google ScholarCross Ref
- The Graph 500 list. http://www.graph500.org.Google Scholar
- L. Gray, A. Kumar, and H. Li. Workload characterization of the SPECpower_ssj2008 benchmark. In SPEC International Performance Evaluation Workshop, June 2008. Google ScholarDigital Library
- The Green500 list: Environmentally responsible supercomputing. http://www.green500.org.Google Scholar
- D. Hackenberg, R. Schöne, D. Molka, M. Müller, and A. Knüpfer. Quantifying power consumption variations of HPC systems using SPEC MPI benchmarks. In International Conference on Energy-Aware High Performance Computing, Sept. 2010.Google ScholarCross Ref
- HPC Challenge benchmark. http://icl.cs.utk.edu/hpcc/.Google Scholar
- C. Hsu, W. Feng, and J. Archuleta. Towards efficient supercomputing: A quest for the right metric. In Workshop on High-Performance, Power-Aware Computing, Apr. 2005. Google ScholarDigital Library
- J. Koomey, S. Berard, M. Sanchez, and H. Wong. Implications of historical trends in the electrical efficiency of computing. IEEE Annals of the History of Computing, 33(3):46--54, July-September 2011. Google ScholarDigital Library
- K.-D. Lange and M. Tricker. The design and development of the server efficiency rating tool (SERT). In International Conference on Performance Engineering, Mar. 2011. Google ScholarDigital Library
- T. Makimoto, K. Eguchi, and M. Yoneyama. The cooler the better: New directions in the nomadic age. IEEE Computer, 34(4):38--42, Apr. 2001. Google ScholarDigital Library
- A. Martin. Towards an energy complexity of computation. Information Processing Letters, 77:181--187, Feb. 2001. Google ScholarDigital Library
- D. Meisner and T. Wenisch. Does low-power design imply energy efficiency for data centers. In International Symposium on Low Power Electronics and Design, Aug. 2011. Google ScholarDigital Library
- S. Microsystems. The SWaP (Space, Watts and Performance) metric, 2007.Google Scholar
- D. Molka, D. Hackenberg, R. Schöne, T. Minartz, and W. Nagel. Flexible workload generation for HPC cluster efficiency benchmarking. In International Conference on Energy-Aware High Performance Computing, Sept. 2011.Google Scholar
- T. Mudge and U. Hölzle. Challenges and opportunities for extremely energy-efficient processors. IEEE Micro, 30(4):20--24, July/August 2010. Google ScholarDigital Library
- M. Müller, M. van Waveren, R. Lieberman, B. Whitney, H. Saito, K. Kumaran, J. Baron, W. Brantley, C. Parrott, T. Elken, H. Feng, and C. Ponder. SPEC MPI2007:an application benchmark suite for parallel systems using MPI. Concurrency and Computation: Practice and Experience, 22(2):191--205, Feb. 2010. Google ScholarDigital Library
- V. Pel. Energy efficiency aspects in Cray supercomputers. In International Conference on Energy-Aware High Performance Computing, Sept. 2010.Google Scholar
- P. Pénzes and A. Martin. Energy-delay efficiency of VLSI computations. In ACM Great Lakes Symposium on VLSI, Apr. 2002. Google ScholarDigital Library
- M. Poess, R. Nambiar, and K. Vaid. Optimizing benchmark configurations for energy efficiency. In International Conference on Performance Engineering, Mar. 2011. Google ScholarDigital Library
- M. Poess, R. Nambiar, K. Vaid, J. J. M. Stephens, K. Huppler, and E. Haines. Energy benchmarks: A detailed analysis. In International Conference on Energy-Efficient Computing and Networking, Apr. 2010. Google ScholarDigital Library
- S. Rivoire, M. Shah, P. Ranganathan, and C. Kozyrakis. JouleSort: A balanced energy-efficiency benchmark. In International Conference on Management of Data, June 2007. Google ScholarDigital Library
- F. Ryckbosch, S. Polfliet, and L. Eeckhout. Trend in server energy-proportionality. IEEE Computer, 44(9):69--72, Sept. 2011. Google ScholarDigital Library
- V. Sarkar. ExaScale software study: Software challenges in extreme scale systems. DARPA IPTO Report, Sept. 2009.Google Scholar
- D. Schall, V. Hoefner, and M. Kern. Towards an enhanced benchmark advocating energy-efficient systems. In TPC Technology Conference on Performance Evaluation & Benchmarking, Aug. 2011. Google ScholarDigital Library
- T. Scogland, B. Subramaniam, and W. Feng. Emerging trends on the evolving Green500: Year three. In Workshop on High-Performance, Power-Aware Computing, May 2011. Google ScholarDigital Library
- SPECpower_ssj2008 Results. http://www.spec.org/power_ssj2008/results/.Google Scholar
- E. Strohmaier. Generalized utility metrics for supercomputers. June 2009.Google Scholar
- The Green Grid. The Green Grid power efficiency metrics: PUE and DCiE, 2007.Google Scholar
- The Green Grid. Recommendations for measuring and reporting overall data center efficiency (version 2), May 2011.Google Scholar
- TOP500 supercomputing sites. http://www.top500.org.Google Scholar
- G. Varsamopoulos and S. Gupta. Energy proportionality and the future: Metrics and directions. In International Workshop on Green Computing, Sept. 2010. Google ScholarDigital Library
Index Terms
- Towards efficient supercomputing: searching for the right efficiency metric
Recommendations
The Green500 list: escapades to exascale
Energy efficiency is now a top priority. The first four years of the Green500 have seen the importance of energy efficiency in supercomputing grow from an afterthought to the forefront of innovation as we approach a point where systems become ...
A power-measurement methodology for large-scale, high-performance computing
ICPE '14: Proceedings of the 5th ACM/SPEC international conference on Performance engineeringImprovement in the energy efficiency of supercomputers can be accelerated by improving the quality and comparability of efficiency measurements. The ability to generate accurate measurements at extreme scale are just now emerging. The realization of ...
Green Supercomputing Comes of Age
Energy-efficient (green) supercomputing has traditionally been viewed as passé, even to the point of public ridicule. But today, it's finally coming into vogue. This article describes the authors' view of this evolution.
Comments