Browse > Article
http://dx.doi.org/10.9708/jksci.2021.26.08.001

Exploiting Hardware Events to Reduce Energy Consumption of HPC Systems  

Lee, Yongho (Dept. of Electrical and Computer Engineering, Sungkyunkwan University)
Kwon, Osang (Dept. of Electrical and Computer Engineering, Sungkyunkwan University)
Byeon, Kwangeun (Dept. of Electrical and Computer Engineering, Sungkyunkwan University)
Kim, Yongjun (Dept. of Electrical and Computer Engineering, Sungkyunkwan University)
Hong, Seokin (Dept. of Semiconductor Systems Engineering, Sungkyunkwan University)
Publication Information
Journal of the Korea Society of Computer and Information / v.26, no.8, 2021 , pp. 1-11 More about this Journal
Abstract
This paper proposes a novel mechanism called Event-driven Uncore Frequency Scaler (eUFS) to improve the energy efficiency of the HPC systems. UFS exploits the hardware events such as LAPI (Last-level Cache Accesses Per Instructions) and CPI (Clock Cycles Per Instruction) to dynamically adjusts the uncore frequency. Hardware events are collected at a reference time period, and the target uncore frequency is determined using the collected event and the previous uncore frequency. Experiments with the NPB benchmarks demonstrate that the eUFS reduces the energy consumption by 6% on average for class C and D NPB benchmarks while it only increases the execution time by 2% on average.
Keywords
HPC; DFS (Dynamic Frequency Scaling); Uncore; Power Management; Low Power;
Citations & Related Records
연도 인용수 순위
  • Reference
1 Marathe, A., Bailey, P. E., Lowenthal, D. K., Rountree, B., Schulz, M., & de Supinski, B. R. "A run-time system for power-constrained HPC applications." International conference on high performance computing, pp. 394-408, July 2015, DOI: 10.1007/978-3-319-20119-1_28   DOI
2 Won, J. Y., Chen, X., Gratz, P., Hu, J., & Soteriou, V. (2014, February). "Up by their bootstraps: Online learning in artificial neural networks for CMP uncore power management." 2014 IEEE 20th International Symposium on High Performance Computer Architecture (HPCA), pp. 308-319, February 2014, DOI: 10.1109/HPCA.2014.6835941   DOI
3 Hackenberg, D., Schone, R., Ilsche, T., Molka, D., Schuchart, J., & Geyer, R. An energy efficiency feature survey of the intel haswell processor. IEEE international parallel and distributed processing symposium workshop, pp. 896-904, May 2015. DOI: 10.1109/IPDPSW.2015.70   DOI
4 Bekele, S. A., Balakrishnan, M., & Kumar, A. "ML guided energy-performance trade-off estimation for uncore frequency scaling." 2019 Spring Simulation Conference (SpringSim), pp. 1-12, April 2019, DOI: 10.23919/SpringSim.2019.8732878   DOI
5 Gholkar, N., Mueller, F., & Rountree, B. "Uncore power scavenger: A runtime for uncore power conservation on hpc systems." Proceedings of the International Conference for High Performance Computing, Networking, Storage and Analysis, pp. 1-23, November 2019, DOI: 10.1145/3295500.3356150   DOI
6 Guide, Part. "Intel® 64 and ia-32 architectures software developer's manual." Volume 3B: System programming Guide, Part 2.11, 2021.
7 Loh, G. "The cost of uncore in throughput-oriented many-core processors." Workshop on Architectures and Languages for Throughput Applications, pp. 1-9, June 2008
8 Sundriyal, V., Sosonkina, M., Westheimer, B., & Gordon, M. "Core and uncore joint frequency scaling strategy. Journal of Computer and Communications", 6(12), pp. 184-201, December 2018, DOI: 10.4236/jcc.2018.612018   DOI
9 Sundriyal, V., Sosonkina, M., Westheimer, B. M., & Gordon, M. "Comparisons of core and uncore frequency scaling modes in quantum chemistry application GAMESS." Proceedings of the High Performance Computing Symposium, pp. 1-11, April 2018, DOI: 10.13140/RG.2.2.15809.45923   DOI
10 Andre, E., Dulong, R., Guermouche, A., & Trahay, F. "DUF: Dynamic Uncore Frequency scaling to reduce power consumption", 2020
11 De Melo, A. C. "The new linux'perf'tools". Slides from Linux Kongress, Vol. 18, pp. 1-42, September 2010.
12 Bailey, D. H., Barszcz, E., Barton, J. T., Browning, D. S., Carter, R. L., Dagum, L., ... & Weeratunga, S. K. "The NAS parallel benchmarks summary and preliminary results." Supercomputing'91: Proceedings of the 1991 ACM/IEEE conference on Supercomputing, pp. 158-165, November 1991, DOI: 10.1145/125826.125925   DOI
13 Hill, D. L., Bachand, D., Bilgin, S., Greiner, R., Hammarlund, P., Huff, T., ... & Safranek, R. "THE UNCORE: A MODULAR APPROACH TO FEEDING THE HIGH-PERFORMANCE CORES". Intel Technology Journal, 14(3), 2010.
14 Guide, Part. "Intel® 64 and ia-32 architectures software developer's manual." Volume 4B: Model- specific registers Part 2.17, 2021.
15 Zhu, G., Han, J., Lee, S., & Son, Y. "An Empirical Evaluation of NVM-aware File Systems on Intel Optane DC Persistent Memory Modules." 2021 International Conference on Information Networking (ICOIN) (pp. 559-564), January 2021, DOI: 10.1109/ICOIN50884.2021.9333911   DOI
16 Cheng, H. Y., Zhan, J., Zhao, J., Xie, Y., Sampson, J., & Irwin, M. J. "Core vs. uncore: The heart of darkness." 2015 52nd ACM/EDAC/IEEE Design Automation Conference (DAC), pp. 1-6, June 2015, DOI: 10.1145/2744769.2647916
17 Takouna, I., Dawoud, W., & Meinel, C. "Analysis and simulation of hpc applications in virtualized data centers." 2012 IEEE International Conference on Green Computing and Communications, pp. 498-507, November 2012, DOI: 10.1109/GreenCom.2012.80   DOI
18 Subramaniam, B., & Feng, W. C. "Towards energy-proportional computing for enterprise-class server workloads." Proceedings of the 4th ACM/SPEC International Conference on Performance Engineering, pp. 15-26, April 2013, DOI: 10.1145/2479871.2479878   DOI
19 Wang, Y., Zhang, W., Hao, M., & Wang, Z.. "Online Power Management for Multi-cores: A Reinforcement Learning Based Approach" IEEE Transactions on Parallel and Distributed Systems, June 2021, DOI: 10.1109/TPDS.2021.3092270   DOI
20 Mankodi, A., Bhatt, A., & Chaudhury, B. "Evaluation of Neural Network Models for Performance Prediction of Scientific Applications." 2020 IEEE REGION 10 CONFERENCE (TENCON), pp. 426-431, November 2020, DOI: 10.1109/TENCON50793.2020.9293788   DOI
21 Freeh, V. W., Kappiah, N., Lowenthal, D. K., & Bletsch, T. K. "Just-in-time dynamic voltage scaling: Exploiting inter-node slack to save energy in mpi programs." Journal of Parallel and Distributed Computing, 68(9), pp. 1175-1185, November 2008 DOI: 10.1016/j.jpdc.2008.04.007   DOI
22 Schone, R., Ilsche, T., Bielert, M., Gocht, A., & Hackenberg, D. "Energy efficiency features of the intel skylake-sp processor and their impact on performance." In 2019 International Conference on High Performance Computing & Simulation (HPCS), pp. 399-406, July 2019, DOI: 10.1109/HPCS48598.2019.9188239   DOI