Exploiting Hardware Events to Reduce Energy Consumption of HPC Systems (하드웨어 이벤트를 활용한 HPC 시스템의 에너지 소비 감소 기법)

LeeYongHo (이용호); KwonOSang (권오상); Kwangeun Byeon (변광은); KIMYONGJUN (김용준); HONG， SEOKIN (홍석인)

doi:10.9708/jksci.2021.26.08.001

Exploiting Hardware Events to Reduce Energy Consumption of HPC Systems

Journal of The Korea Society of Computer and Information
Abbr : JKSCI
2021, 26(8), pp.1-11
DOI : 10.9708/jksci.2021.26.08.001
Publisher : The Korean Society Of Computer And Information
Research Area : Engineering > Computer Science
Received : July 28, 2021
Accepted : August 23, 2021
Published : August 31, 2021

LeeYongHo ¹, KwonOSang ¹, Kwangeun Byeon ¹, KIMYONGJUN ¹, HONG， SEOKIN ¹

¹성균관대학교

Accredited

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

Citation status

* References for papers published after 2023 are currently being built.

[confproc] Hackenberg, D. / 2015 / An energy efficiency feature survey of the intel haswell processor / IEEE international parallel and distributed processing symposium workshop : 896~904

[confproc] Gholkar, N. / 2019 / 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 : 1~23

[other] Guide, Part / 2021 / Intel® 64 and ia-32 architectures software developer’s manual / Volume 3B: System programming Guide, Part 2.11

[confproc] Loh, G. / 2008 / The cost of uncore in throughput-oriented many-core processors / Workshop on Architectures and Languages for Throughput Applications : 1~9

[confproc] Marathe, A / 2015 / A run-time system for power-constrained HPC applications / International conference on high performance computing : 394~408

[journal] Vincent W. Freeh / 2008 / Just-in-time dynamic voltage scaling: Exploiting inter-node slack to save energy in MPI programs / Journal of Parallel and Distributed Computing / Elsevier BV 68(9) : 1175~1185

[confproc] Bekele, S. A. / 2019 / ML guided energy-performance trade-off estimation for uncore frequency scaling / 2019 Spring Simulation Conference : 1~12

[confproc] Won, J. Y. / 2014 / 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) : 308~319

[journal] Vaibhav Sundriyal / 2018 / Core and Uncore Joint Frequency Scaling Strategy / Journal of Computer and Communications / Scientific Research Publishing, Inc. 06(12) : 184~201

[confproc] Sundriyal, V. / 2018 / Comparisons of core and uncore frequency scaling modes in quantum chemistry application GAMESS / Proceedings of the High Performance Computing Symposium : 1~11

[report] André, E. / 2020 / DUF:Dynamic Uncore Frequency scaling to reduce power consumption

[confproc] Bailey, D. H. / 1991 / The NAS parallel benchmarks summary and preliminary results / Supercomputing'91:Proceedings of the 1991 ACM/IEEE conference on Supercomputing : 158~165

[journal] Hill, D. L. / 2010 / THE UNCORE: A MODULAR APPROACH TO FEEDING THE HIGH-PERFORMANCE CORES / Intel Technology Journal 14(3)

[other] Guide, Part / 2021 / Intel® 64 and ia-32 architectures software developer’s manual / Volume 4B: Model- specific registers Part 2.17

[journal] De Melo, A. C. / 2010 / The new linux’perf’tools / Slides from Linux Kongress 18 : 1~42

[confproc] Zhu, G. / 2021 / An Empirical Evaluation of NVM-aware File Systems on Intel Optane DC Persistent Memory Modules / 2021 International Conference on Information Networking (ICOIN) : 559~564

[confproc] Schöne, R. / 2019 / Energy efficiency features of the intel skylake-sp processor and their impact on performance / 2019 International Conference on High Performance Computing & Simulation (HPCS) : 399~406

[confproc] Cheng, H. Y. / 2015 / Core vs. uncore: The heart of darkness / 2015 52nd ACM/EDAC/IEEE Design Automation Conference (DAC) : 1~6

[confproc] Subramaniam, B. / 2013 / Towards energy-proportional computing for enterprise-class server workloads / Proceedings of the 4th ACM/SPEC International Conference on Performance Engineering : 15~26

[journal] Wang, Y. / 2021 / Online Power Management for Multi-cores: A Reinforcement Learning Based Approach / IEEE Transactions on Parallel and Distributed Systems

[confproc] Takouna, I. / 2012 / Analysis and simulation of hpc applications in virtualized data centers / 2012 IEEE International Conference on Green Computing and Communications : 498~507

[confproc] Mankodi, A. / 2020 / Evaluation of Neural Network Models for Performance Prediction of Scientific Applications / 2020 IEEE REGION 10 CONFERENCE (TENCON) : 426~431

This paper was written with support from the National Research Foundation of Korea.

KJCKorea
Journal Central

Journal of The Korea Society of Computer and Information 2023 KCI Impact Factor : 0.65

Exploiting Hardware Events to Reduce Energy Consumption of HPC Systems

ABSTRACT

KEYWORDS

Citation status

* References for papers published after 2023 are currently being built.

Journal of The Korea Society of Computer and Information 2023 KCI Impact Factor : 0.65

Exploiting Hardware Events to Reduce Energy Consumption of HPC Systems

ABSTRACT

KEYWORDS

Statistics

Tools

Issue List

Citation status

KCI Citation Counts (0)

REFERENCES (22) * References for papers published after 2023 are currently being built.

Search PDF

Citation

* References for papers published after 2023 are currently being built.