Cited By
View all- Patel PGong ZRizvi SChoukse EMisra PAnderson TSriraman A(2023)Towards Improved Power Management in Cloud GPUsIEEE Computer Architecture Letters10.1109/LCA.2023.327865222:2(141-144)Online publication date: 1-Jul-2023
Predict; Don't React for Enabling Efficient Fine-Grain DVFS in GPUs
Authors: 
Srikant Bharadwaj, Shomit Das, Kaushik Mazumdar, Bradford M. Beckmann, Stephen KosonockyAuthors Info & Claims
Srikant Bharadwaj, Shomit Das, Kaushik Mazumdar, Bradford M. Beckmann, Stephen KosonockyAuthors Info & ClaimsPages 253 - 267
Abstract
With the continuous improvement of on-chip integrated voltage regulators (IVRs) and fast, adaptive frequency control, dynamic voltage-frequency scaling (DVFS) transition times have shrunk from the microsecond to the nanosecond regime, providing immense opportunity to improve energy efficiency. The key to unlocking the continued improvement in V/f circuit technology is the creation of new, smarter DVFS mechanisms that better adapt to rapid fluctuations in workload demand.
It is particularly important to optimize fine-grain DVFS mechanisms for graphics processing units (GPUs) as the chips become ever more important workhorses in the datacenter. However, GPU's massive amount of thread-level parallelism makes it uniquely difficult to determine the optimal V/f state at run-time. Existing solutions---mostly designed for single-threaded CPUs and longer time scales---fail to consider the seemingly chaotic, highly varying nature of GPU workloads at short time scales.
This paper proposes a novel prediction mechanism, PCSTALL, that is tailored for emerging DVFS capabilities in GPUs and achieves near-optimal energy efficiency. Using the insights from our fine-grained workload analysis, we propose a wavefront-level program counter (PC) based DVFS mechanism that improves program behavior prediction accuracy by 32% on average as compared to the best performing prior predictor for a wide set of GPU applications at 1μs DVFS time epochs. Compared to the current state-of-art, our PC-based technique achieves 19% average improvement when optimized for Energy-Delay2 Product (ED2P) at 50μs time epochs, reaching 32% when operated with 1μs DVFS technologies.
References
[1]
S. Kaxiras and M. Martonosi, "Computer Architecture Techniques for Power-Efficiency," Synthesis Lectures on Computer Architecture, vol. 3, no. 1, pp. 1--207, 2008.
[2]
S. Eyerman and L. Eeckhout, "Fine-grained DVFS using on-chip regulators," ACM Transactions on Architecture and Code Optimization, vol. 8, no. 1, pp. 1--24, 2011.
[3]
W. Kim, M. S. Gupta, G.-Y. Wei and D. M. Brooks, "System level analysis of fast, per-core DVFS using on-chip switching regulators," 2008. [Online]. Available: http://eecs.harvard.edu/~dbrooks/kim2008_hpca.pdf. [Accessed 25 6 2019].
[4]
Y. Bai, V. W. Lee and E. Ipek, "Voltage Regulator Efficiency Aware Power Management," Sigplan Notices, vol. 45, no. 1, pp. 825--838, 2017.
[5]
A. Paul, S. P. Park, S. Dinesh, Y. M. Kim, N. Borkar, U. R. Karpuzcu and C. H. Kim, "System-Level Power Analysis of a Multicore Multipower Domain Processor With ON-Chip Voltage Regulators," IEEE Transactions on Very Large Scale Integration (VLSI) Systems, vol. 24, pp. 3468--3476, 2016.
[6]
E. A. Burton, G. Schrom, F. Paillet, J. P. Douglas, W. J. Lambert, K. Radhakrishnan and M. J. Hill, "FIVR --- Fully integrated voltage regulators on 4th generation Intel® CoreTM SoCs," 2014. [Online]. Available: https://ieeexplore.ieee.org/document/6803344. [Accessed 21 11 2019].
[7]
M. Fojtik, B. Keller, A. Klinefelter, N. Pinckney, S. G. Tell, B. Zimmer, T. Raja, K. Zhou, W. J. Dally and B. Khailany, "A Fine-Grained GALS SoC with Pausible Adaptive Clocking in 16 nm FinFET," in 2019 25th IEEE International Symposium on Asynchronous Circuits and Systems (ASYNC), 2019.
[8]
X. Sun, A. Boora, W. Zhang, V. R. Pamula and V. Sathe, "14.5 A 0.6-to-1.1V Computationally Regulated Digital LDO with 2.79-Cycle Mean Settling Time and Autonomous Runtime Gain Tracking in 65nm CMOS," IEEE International Solid- State Circuits Conference - (ISSCC), 2019.
[9]
Y. Okuma, K. Ishida, Y. Ryu, X. Zhang, P.-H. Chen, K. Watanabe, M. Takamiya and T. Sakurai, "0.5-V input digital LDO with 98.7% current efficiency and 2.7-μA quiescent current in 65nm CMOS," 2010. [Online]. Available: http://lowpower.iis.u-tokyo.ac.jp/paper/2010_22.pdf. [Accessed 25 11 2019].
[10]
R. Miftakhutdinov, E. Ebrahimi and Y. N. Patt, "Predicting Performance Impact of DVFS for Realistic Memory Systems," IEEE Micro, pp. 155--165, 2012.
[11]
P. Paternoster, A. Maki, A. Hernandez, M. Grossman, M. Lau, D. Sutherland and A. Mathad, "XBOX Series X: A Next-Generation Gaming Console SoC," in 2021 IEEE International Solid- State Circuits Conference (ISSCC), 2021.
[12]
P. A. Meinerzhagen, C. Tokunaga, A. Malavasi, V. Vaidya, A. Mendon, D. A. Mathaikutty, J. P. Kulkarni, C. Augustine, M. Cho, S. T. Kim, G. E. Matthew, R. Jain, J. F. Ryan, C.-C. Peng, S. Paul, S. R. Vangal, B. P. Esparza, L. Cuellar, M. Woodman, B. Iyer, S. Maiyuran, G. N. Chinya, C. Zou, Y. Liao, K. Ravichandran, H. Wang, M. M. Khellah, J. W. Tschanz and V. De, "An energy-efficient graphics processor featuring fine-grain DVFS with integrated voltage regulators, execution-unit turbo, and retentive sleep in 14nm tri-gate CMOS," 2018. [Online]. Available: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=8310172. [Accessed 16 8 2019].
[13]
T. Singh, A. Schaefer, S. Rangarajan, D. John, C. Henrion, R. Schreiber, M. Rodriguez, S. Kosonocky, S. D. Naffziger and A. Novak, "Zen: An Energy-Efficient High-Performance X86 Core," IEEE Journal of Solid-state Circuits, vol. 53, no. 1, pp. 102--114, 2018.
[14]
T. Burd, N. Beck, S. White, M. Paraschou, N. Kalyanasundharam, G. Donley, A. D. Smith, L. D. Hewitt and S. D. Naffziger, ""Zeppelin": An SoC for Multichip Architectures," IEEE Journal of Solid-state Circuits, vol. 54, no. 1, pp. 133--143, 2019.
[15]
D. Bouvier, J. Gibney, A. Branover and S. Arora, "AMD Raven-Ridge APU: Delivering a new level of visual performance in an SoC," [Online]. Available: https://www.hotchips.org/hc30/1conf/1.05_AMD_APU_AMD_Raven_HotChips30_Final.pdf.
[16]
B. Zimmer, Y. Lee, A. Puggelli, J. Kwak, R. Jevtic, B. Keller, S. Bailey, M. Blagojevic, P.-F. Chiu, H.-P. Le, P.-H. Chen, N. Sutardja, R. Avizienis, A. Waterman, B. Richards, P. Flatresse, E. Alon, K. Asanovic and B. Nikolic, "A RISC-V vector processor with tightly-integrated switched-capacitor DC-DC converters in 28nm FDSOI," 2015. [Online]. Available: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=7231305. [Accessed 12 2 2020].
[17]
R. Muthukaruppan, T. Mahajan, H. Krishnamurthy, S. Mangal, A. Dhanshekhar, R. Ghayal and V. De, "A digitally controlled linear regulator for per-core wide-range DVFS of atomTM cores in 14nm tri-gate CMOS featuring non-linear control, adaptive gain and code roaming," in ESSCIRC 2017 - 43rd IEEE European Solid State Circuits Conference, Leuven, 2017.
[18]
S. J. Kim, D. Kim, H. Ham, J. Kim and M. Seok, "A 67.1-ps FOM, 0.5-V-Hybrid Digital LDO With Asynchronous Feedforward Control Via Slope Detection and Synchronous PI With State-Based Hysteresis Clock Switching," 2018. [Online]. Available: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=8490730. [Accessed 12 2 2020].
[19]
H. Li, X. Wang, J. Xu, Z. Wang, R. K. V. Maeda, Z. Wang, P. Yang, L. H. K. Duong and Z. W. Wang, "Energy-Efficient Power Delivery System Paradigms for Many-Core Processors," IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems, vol. 36, no. 3, pp. 449--462, 2017.
[20]
R. Nath and D. M. Tullsen, "The CRISP performance model for dynamic voltage and frequency scaling in a GPGPU," in 2015 48th Annual IEEE/ACM International Symposium on Microarchitecture (MICRO), 2015.
[21]
X. Mei, L. S. Yung, K. Zhao and X. Chu, "A measurement study of GPU DVFS on energy conservation," 2013. [Online]. Available: http://comp.hkbu.edu.hk/~chxw/papers/hotpower_2013.pdf. [Accessed 15 8 2019].
[22]
J. a. B. A. Leng, R. Bertran, P. Bose, Y. Zu and V. J. Reddi, "Predictive Guardbanding: Program-Driven Timing Margin Reduction for GPUs," IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems, vol. 40, no. 1, pp. 171--184, 2021.
[23]
S. T. Kim, Y.-C. Shih, K. Mazumdar, R. Jain, J. F. Ryan, C. Tokunaga, C. Augustine, J. P. Kulkarni, K. Ravichandran, J. W. Tschanz, M. M. Khellah and V. De, "Enabling Wide Autonomous DVFS in a 22 nm Graphics Execution Core Using a Digitally Controlled Fully Integrated Voltage Regulator," IEEE Journal of Solid-state Circuits, vol. 51, no. 1, pp. 18--30, 2016.
[24]
G. Keramidas, V. Spiliopoulos and S. Kaxiras, "Interval-based models for runtime DVFS orchestration in superscalar processors," 2010. [Online]. Available: http://pages.cs.wisc.edu/~kaxiras/papers/cf10_dvfs_model.pdf. [Accessed 25 6 2019].
[25]
C. Isci, G. Contreras and M. Martonos, Live, Runtime Phase Monitoring and Prediction on Real Systems with Application to Dynamic Power Management, 2006, p. 359--370.
[26]
K. Straube, J. Lowe-Power, C. Nitta, M. Farrens and V. Akella, "Improving Provisioned Power Efficiency in HPC Systems with GPU-CAPP," 2018 IEEE 25th International Conference on High Performance Computing (HiPC), 2018.
[27]
C. C. Nugteren, G. G.-J. v. d. Braak and H. Corporaal, "Roofline-aware DVFS for GPUs," 2014. [Online]. Available: http://ece.neu.edu/groups/nucar/nucartalks/roofline-aware_dvfs_for_gpus.pdf. [Accessed 15 8 2019].
[28]
A. Mishra and N. Khare, "Analysis of DVFS Techniques for Improving the GPU Energy Efficiency," 2015. [Online]. Available: http://file.scirp.org/pdf/ojee_2015121415504865.pdf. [Accessed 15 8 2019].
[29]
Z. Tang, Y. Wang, Q. Wang and X. Chu, "The Impact of GPU DVFS on the Energy and Performance of Deep Learning: An Empirical Study," in Proceedings of the Tenth ACM International Conference on Future Energy Systems, Phoenix, 2019.
[30]
G. Antoniou, H. Volos, D. B. Bartolini, T. Rollet, Y. Sazeides and J. H. Yahya, "AgilePkgC: An Agile System Idle State Architecture for Energy Proportional Datacenter Servers," arXiv, 2022.
[31]
B. Dutta, V. Adhinarayanan and W.-c. Feng, "GPU power prediction via ensemble machine learning for DVFS space exploration," 2018. [Online]. Available: https://vtechworks.lib.vt.edu/handle/10919/81997. [Accessed 15 8 2019].
[32]
K. Fan, B. Cosenza and B. Juurlink, "Predictable GPUs Frequency Scaling for Energy and Performance," in Proceedings of the 48th International Conference on Parallel Processing, Kyoto, Japan, 2019.
[33]
S. Eyerman and L. Eeckhout, "A Counter Architecture for Online DVFS Profitability Estimation," IEEE Transactions on Computers, vol. 59, no. 11, pp. 1576--1583, 2010.
[34]
B. Rountree, D. K. Lowenthal, M. Schulz and B. R. d. Supinski, "Practical performance prediction under Dynamic Voltage Frequency Scaling," 2011. [Online]. Available: http://yadda.icm.edu.pl/yadda/element/bwmeta1.element.ieee-000006008553. [Accessed 25 6 2019].
[35]
C. Isci, A. Buyuktosunoglu, C.-Y. Chen, P. Bose and M. Martonosi, "An Analysis of Efficient Multi-Core Global Power Management Policies: Maximizing Performance for a Given Power Budget," IEEE Micro, pp. 347--358, 2006.
[36]
H. Hoffmann and M. Maggio, "PCP: A Generalized Approach to Optimizing Performance Under Power Constraints through Resource Management," 2014. [Online]. Available: https://usenix.org/system/files/conference/icac14/icac14-paper-hoffman.pdf. [Accessed 23 7 2019].
[37]
S. Park, J. Park, D. Shin, Y. Wang, Q. Xie, M. Pedram and N. Chang, "Accurate Modeling of the Delay and Energy Overhead of Dynamic Voltage and Frequency Scaling in Modern Microprocessors," IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems, vol. 32, no. 5, pp. 695--708, 2013.
[38]
V. Spiliopoulos, S. Kaxiras and G. Keramidas, "Green governors: A framework for Continuously Adaptive DVFS," 2011. [Online]. Available: http://diva-portal.org/smash/record.jsf?pid=diva2:474791. [Accessed 25 6 2019].
[39]
M. Weiser, B. B. Welch, A. J. Demers and S. Shenker, "Scheduling for reduced CPU energy," 1994. [Online]. Available: https://link.springer.com/chapter/10.1007/978-0-585-29603-6_17. [Accessed 25 6 2019].
[40]
P. Zou, A. Li, K. Barker and R. Ge, "Indicator-Directed Dynamic Power Management for Iterative Workloads on GPU-Accelerated Systems," in 2020 20th IEEE/ACM International Symposium on Cluster, Cloud and Internet Computing (CCGRID), 2020.
[41]
A. Zou, J. Leng, Y. Zu, T. Tong, V. J. Reddi, D. M. Brooks, G.-Y. Wei and X. Zhang, "Ivory: Early-Stage Design Space Exploration Tool for Integrated Voltage Regulators," 2017. [Online]. Available: https://dl.acm.org/citation.cfm?id=3062268. [Accessed 23 7 2019].
[42]
A. Paul, S. P. Park, D. Somasekhar, Y. M. Kim, N. Borkar, U. R. Karpuzcu and C.H.Kim, "System-Level Power Analysis of a Multicore Multipower Domain Processor With ON-Chip Voltage Regulators," IEEE Transactions on Very Large Scale Integration (VLSI) Systems, 2016.
[43]
J. Haj-Yahya, M. Alser, J. S. Kim, L. Orosa, E. Rotem, A. Mendelson, A. Chattopadhyay and O. Mutlu, "FlexWatts: A Power- and Workload-Aware Hybrid Power Delivery Network for Energy-Efficient Microprocessors," arXiv, 2020.
[44]
M. Sankarasubramanian, K. Radhakrishnan, Y. Min, W. Lambert, M. J. Hill, A. Dani, R. Mesch, L. Wojewoda, J. Chavarria and A. Augustine, "Magnetic Inductor Arrays for Intel® Fully Integrated Voltage Regulator (FIVR) on 10th generation Intel® CoreTM SoCs," in 2020 IEEE 70th Electronic Components and Technology Conference (ECTC), Orlando, 2020.
[45]
C. Schaef, K. Radhakrishnan, K. Ravichandran, J. W. Tschanz, V. De, N. Desai, H. K. Krishnamurthy, X. Liu, K. Z. Ahmed, S. Kim, S. Weng, H. T. Do and W. J. Lambert, "A Light-Load Efficient Fully Integrated Voltage Regulator in 14-nm CMOS With 2.5-nH Package-Embedded Air-Core Inductors," IEEE Journal of Solid-state Circuits, vol. 54, no. 12, pp. 3316--3325, 2019.
[46]
G. Yan, Y. Li, Y. Han, X. Li, M. Guo and X. Liang, "AgileRegulator: A hybrid voltage regulator scheme redeeming dark silicon for power efficiency in a multicore architecture," 2012. [Online]. Available: http://cs.sjtu.edu.cn/~guo-my/pdf/conferences/c130.pdf. [Accessed 5 7 2019].
[47]
Q. Deng, D. Meisner, A. Bhattacharjee, T. F. Wenisch and R. Bianchini, "CoScale: Coordinating CPU and Memory System DVFS in Server Systems," IEEE Micro, pp. 143--154, 2012.
[48]
S. Akram, J. B. Sartor and L. Eeckhout, "DVFS performance prediction for managed multithreaded applications," 2016. [Online]. Available: https://biblio.ugent.be/publication/7245653. [Accessed 23 7 2019].
[49]
H. Zhang and H. Hoffmann, "Maximizing Performance Under a Power Cap: A Comparison of Hardware, Software, and Hybrid Techniques," Sigplan Notices, vol. 44, no. 2, pp. 545--559, 2016.
[50]
H. Hoffmann and M. Maggio, "PCP: A Generalized Approach to Optimizing Performance Under Power Constraints through Resource Management," 2014. [Online]. Available: https://lup.lub.lu.se/search/publication/44544051-23aa-4be2-904f-b780181c3f90. [Accessed 5 7 2019].
[51]
C. C. Lin, C. J. Chang, Y. C. Syu, J. J. Wu, P. Liu, P. W. Cheng and W. T. Hsu, "An Energy-Efficient Task Scheduler for Multi-core Platforms with Per-core DVFS Based on Task Characteristics," 2014. [Online]. Available: http://ieeexplore.ieee.org/document/6957247. [Accessed 5 7 2019].
[52]
M. Curtis-Maury, J. Dzierwa, C. D. Antonopoulos and D. S. Nikolopoulos, "Online power-performance adaptation of multithreaded programs using hardware event-based prediction," 2006. [Online]. Available: http://people.cs.vt.edu/~dsn/papers/ics06.pdf. [Accessed 5 7 2019].
[53]
B. Su, J. Gu, L. Shen, W. Huang, J. L. Greathouse and Z. Wang, "PPEP: Online Performance, Power, and Energy Prediction Framework and DVFS Space Exploration," IEEE Micro, pp. 445--457, 2014.
[54]
A. Gendler, E. Knoll and Y. Sazeides, "I-DVFS: Instantaneous Frequency Switch During Dynamic Voltage and Frequency Scaling," in IEEE Micro, 2021.
[55]
J. Haj-Yahya, M. Alser, J. Kim, A. G. Yaglikçi, N. Vijaykumar, E. Rotem and O. Mutlu, "SysScale: Exploiting Multi-domain Dynamic Voltage and Frequency Scaling for Energy Efficient Mobile Processors," in Proceedings of the ACM/IEEE 47th Annual International Symposium on Computer Architecture, Virtual Event, 2020.
[56]
A. Miyoshi, C. Lefurgy, E. Van Hensbergen, R. Rajamony and R. Rajkumar, "Critical Power Slope: Understanding the Runtime Effects of Frequency Scaling," in Proceedings of the 16th International Conference on Supercomputing, New York, New York, USA, 2002.
[57]
Q. Wu, M. Martonosi, D. W. Clark, V. J. Reddi, D. A. Connors, Y. Wu, J. Lee and D. M. Brooks, "A Dynamic Compilation Framework for Controlling Microprocessor Energy and Performance," IEEE Micro, pp. 271--282, 2005.
[58]
Y. Sazeides, R. Kumar, D. Tullsen and T. Constantinou, "The Danger of Interval-Based Power Efficiency Metrics: When Worst Is Best," in IEEE Computer Architecture Letters, 2005.
[59]
S. Akram, J. B. Sartor and L. Eeckhout, "DEP+BURST: Online DVFS Performance Prediction for Energy-Efficient Managed Language Execution," IEEE Transactions on Computers, vol. 66, no. 4,, pp. pp. 601--615, 1 April 2017.
[60]
C. Isci, A. Buyuktosunoglu and M. Martonosi, "Long-term workload phases: duration predictions and applications to DVFS," IEEE Micro, vol. 25, no. 5, pp. 39--51, 2005.
[61]
W. L. Bircher and L. K. John, "Predictive power management for multi-core processors,", 2010. [Online]. Available: https://link.springer.com/chapter/10.1007/978-3-642-24322-6_21. [Accessed 21 11 2019].
[62]
A. Smith and N. James, "AMD InstinctTM MI200 Series Accelerator and Node Architectures," in 2022 IEEE Hot Chips 34 Symposium (HCS), 2022.
[63]
R. Thomas, N. Sedaghati and R. Teodorescu, "EmerGPU: Understanding and mitigating resonance-induced voltage noise in GPU architectures," 2016. [Online]. Available: https://ieeexplore.ieee.org/document/7482076. [Accessed 24 11 2021].
[64]
C. Isci and M. Martonosi, "Phase characterization for power: evaluating control-flow-based and event-counter-based techniques," 2006. [Online]. Available: http://parapet.ee.princeton.edu/papers/canturk-hpca2006.pdf. [Accessed 15 8 2019].
[65]
O. Khan and S. Kundu, "Microvisor: a runtime architecture for thermal management in chip multiprocessors," 2011. [Online]. Available: https://link.springer.com/chapter/10.1007/978-3-642-24568-8_5. [Accessed 20 11 2019].
[66]
S. Srinivasan, R. Kumar and S. Kundu, "Program phase duration prediction and its application to fine-grain power management," 2013. [Online]. [Accessed 20 11 2019].
[67]
F. Xie, M. Martonosi and S. Malik, "Compile-time dynamic voltage scaling settings: opportunities and limits," Sigplan Notices, vol. 38, no. 5, pp. 49--62, 2003.
[68]
A. Gutierrez, B. M. Beckmann, A. Dutu, J. Gross, M. LeBeane, J. Kalamatianos, O. Kayiran, M. Poremba, B. Potter, S. Puthoor, M. D. Sinclair, M. Wyse, J. Yin, X. Zhang, A. Jain and T. G. Rogers, "Lost in Abstraction: Pitfalls of Analyzing GPUs at the Intermediate Language Level," 2018. [Online]. Available: https://ieeexplore.ieee.org/document/8327041. [Accessed 30 7 2019].
[69]
N. L. Binkert, B. M. Beckmann, G. Black, S. K. Reinhardt, A. G. Saidi, A. Basu, J. Hestness, D. R. Hower, T. Krishna, S. Sardashti, R. Sen, K. Sewell, M. Shoaib, N. Vaish, M. D. Hill and D. A. Wood, "The gem5 simulator," ACM Sigarch Computer Architecture News, vol. 39, no. 2, pp. 1--7, 2011.
[70]
S. Bharadwaj, J. Yin, B. Beckmann and T. Krishna, "Kite: A Family of Heterogeneous Interposer Topologies Enabled via Accurate Interconnect Modeling," in 57th ACM/IEEE Design Automation Conference (DAC), San Francisco, CA, US, 2020.
[71]
J. L. Greathouse and G. H. Loh, "Machine learning for performance and power modeling of heterogeneous systems," In Proceedings of the International Conference on Computer-Aided Design (ICCAD '18). ACM, New York, NY, USA, Article 47, 6pages, 2018.
[72]
I. Karlin, J. Keasler and J. R. Neely, "LULESH 2.0 Updates and Changes," 2013. [Online]. Available: https://codesign.llnl.gov/pdfs/lulesh2.0_changes.pdf. [Accessed 2 8 2019].
[73]
BAIDU Research, "DeepBench," [Online]. Available: https://github.com/baidu-research/DeepBench.
[74]
S. Dong and D. R. Kaeli, "DNNMark: A Deep Neural Network Benchmark Suite for GPUs," 2017. [Online]. Available: https://dl.acm.org/citation.cfm?id=3038239. [Accessed 2 8 2019].
[75]
P. Shah, R. G. Shenoy, V. Srinivasan, P. Bose and A. Buyuktosunoglu, "TokenSmart: Distributed, Scalable Power Management in the Many-Core Era," in IEEE Computer Architecture Letters, 2021.
[76]
C. Isci, G. Contreras and M. Martonosi, "Live, Runtime Phase Monitoring and Prediction on Real Systems with Application to Dynamic Power Management," IEEE Micro, pp. 359--370, 2006.
[77]
K. Choi, R. Soma and M. Pedram, "Fine-grained dynamic voltage and frequency scaling for precise energy and performance trade-off based on the ratio of off-chip access to on-chip computation times," 2004. [Online]. Available: http://sportlab.usc.edu/~kihwan/fg-dvfs.pdf. [Accessed 5 7 2019].
[78]
V. Spiliopoulos, A. Bagdia, A. Hansson, P. Aldworth and S. Kaxiras, "Introducing DVFS-Management in a Full-System Simulator," 2013. [Online]. Available: http://it.uu.se/katalog/vassp447/gem5_dvfs.pdf. [Accessed 30 7 2019].
[79]
Q. Wu, V. J. Reddi, Y. Wu, J. Lee, D. Connors, D. Brooks, M. Martonosi, D. W. Clark and yes, A Dynamic Compilation Framework for Controlling Microprocessor Energy and Performance, 2005, p. 271--282.
[80]
Z. Toprak-Deniz, M. A. Sperling, J. F. Bulzacchelli, G. S. Still, R. Kruse, S. Kim, D. W. Boerstler, T. Gloekler, R. Robertazzi, K. Stawiasz, T. Diemoz, G. English, D. T. Hui, P. H. Muench and J. Friedrich, "5.2 Distributed system of digitally controlled microregulators enabling per-core DVFS for the POWER8 TM microprocessor," 2014. [Online]. Available: http://ieeexplore.ieee.org/document/6757354. [Accessed 25 11 2019].
[81]
H. Li, J. Xu, Z. Wang, R. K. V. Maeda, P. Yang and Z. Tian, "Workload-Aware Adaptive Power Delivery System Management for Many-Core Processors," IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems, vol. 37, no. 10, pp. 2076--2086, 2018.
[82]
A. Jog, O. Kayiran, A. Pattnaik, M. Kandemir, O. Mutlu, R. Iyer and C. R. Das, "Exploiting Core Criticality for Enhanced GPU Performance," Sigmetrics Performance Evaluation Review, vol. 44, no. 1, pp. 351--363, 2016.
[83]
AMD, "Polaris Whitepaper".
[84]
L. Wang, M. Jahre, A. Adileho and L. Eeckhout, "MDM: The GPU Memory Divergence Model," in International Symposium on Microarchitecture(MICRO), 2020.
Index Terms
- Predict; Don't React for Enabling Efficient Fine-Grain DVFS in GPUs
Recommendations
Roofline-aware DVFS for GPUs
ADAPT '14: Proceedings of International Workshop on Adaptive Self-tuning Computing SystemsGraphics processing units (GPUs) are becoming increasingly popular for compute workloads, mainly because of their large number of processing elements and high-bandwidth to off-chip memory. The roofline model captures the ratio between the two (the ...
On the Efficacy of a Fused CPU+GPU Processor (or APU) for Parallel Computing
SAAHPC '11: Proceedings of the 2011 Symposium on Application Accelerators in High-Performance ComputingThe graphics processing unit (GPU) has made significant strides as an accelerator in parallel computing. However, because the GPU has resided out on PCIe as a discrete device, the performance of GPU applications can be bottlenecked by data transfers ...
Optimizing Memory Efficiency for Convolution Kernels on Kepler GPUs
DAC '17: Proceedings of the 54th Annual Design Automation Conference 2017Convolution is a fundamental operation in many applications, such as computer vision, natural language processing, image processing, etc. Recent successes of convolutional neural networks in various deep learning applications put even higher demand on ...
Comments
Information & Contributors
Information
Published In
March 2023
430 pages
ISBN:9798400703942
DOI:10.1145/3623278
- Chair:
- Tor Aamodt,
- Program Chair:
- Michael M Swift,
- Program Co-chair:
- Natalie Enright Jerger
Copyright © 2023 ACM.
Publication rights licensed to ACM. ACM acknowledges that this contribution was authored or co-authored by an employee, contractor or affiliate of the United States government. As such, the Government retains a nonexclusive, royalty-free right to publish or reproduce this article, or to allow others to do so, for Government purposes only.
Sponsors
In-Cooperation
Publisher
Association for Computing Machinery
New York, NY, United States
Publication History
Published: 07 February 2024
Check for updates
Author Tags
Qualifiers
- Research-article
Conference
ASPLOS '23: 28th ACM International Conference on Architectural Support for Programming Languages and Operating Systems, Volume 4
March 25 - 29, 2023
BC, Vancouver, Canada
Acceptance Rates
Overall Acceptance Rate 535 of 2,713 submissions, 20%
Contributors
Other Metrics
Bibliometrics & Citations
Bibliometrics
Article Metrics
- View Citations1Total Citations
- 446Total Downloads
- Downloads (Last 12 months)446
- Downloads (Last 6 weeks)111
Reflects downloads up to 03 Oct 2024
Other Metrics
Citations
Cited By
View all- Patel PGong ZRizvi SChoukse EMisra PAnderson TSriraman A(2023)Towards Improved Power Management in Cloud GPUsIEEE Computer Architecture Letters10.1109/LCA.2023.327865222:2(141-144)Online publication date: 1-Jul-2023
View Options
Get Access
Login options
Check if you have access through your login credentials or your institution to get full access on this article.
Sign in