GPU power prediction via ensemble machine learning for DVFS space exploration

A software-based approach to achieve high performance within a power budget often involves dynamic voltage and frequency scaling (DVFS). Thus, accurately predicting the power consumption of an application at different DVFS levels (or more generally, different processor configurations) is paramount for the energy-efficient functioning of a high-performance computing (HPC) system. The increasing prevalence of graphics processing units (GPUs) in HPC systems presents new challenges in power management, and machine learning presents an unique way to improve the software-based power management of these systems. As such, we explore the problem of GPU power prediction at different DVFS states via machine learning. Specifically, we propose a new ensemble technique that incorporates three machine-learning techniques --- sequential minimal optimization regression, simple linear regression, and decision tree --- to reduce the mean absolute error (MAE) to 3.5%.

[1]  Shiliang Sun Traffic flow forecasting based on multitask ensemble learning , 2009, GEC '09.

[2]  Derek Chiou,et al.  GPGPU performance and power estimation using machine learning , 2015, 2015 IEEE 21st International Symposium on High Performance Computer Architecture (HPCA).

[3]  Karthick Rajamani,et al.  A performance-conserving approach for reducing peak power consumption in server systems , 2005, ICS '05.

[4]  Wu-chun Feng,et al.  An automated framework for characterizing and subsetting GPGPU workloads , 2016, 2016 IEEE International Symposium on Performance Analysis of Systems and Software (ISPASS).

[5]  Xiaohan Ma,et al.  Statistical Power Consumption Analysis and Modeling for GPU-based Computing , 2011 .

[6]  Wu-chun Feng,et al.  Online Power Estimation of Graphics Processing Units , 2016, 2016 16th IEEE/ACM International Symposium on Cluster, Cloud and Grid Computing (CCGrid).

[7]  Wu-chun Feng,et al.  Statistical Power and Performance Modeling for Optimizing the Energy Efficiency of Scientific Computing , 2010, 2010 IEEE/ACM Int'l Conference on Green Computing and Communications & Int'l Conference on Cyber, Physical and Social Computing.

[8]  Xiaorui Wang,et al.  Power capping: a prelude to power shifting , 2008, Cluster Computing.

[9]  Sherief Reda,et al.  Pack & Cap: Adaptive DVFS and thread packing under power caps , 2011, 2011 44th Annual IEEE/ACM International Symposium on Microarchitecture (MICRO).

[10]  Rahul Khanna,et al.  RAPL: Memory power estimation and capping , 2010, 2010 ACM/IEEE International Symposium on Low-Power Electronics and Design (ISLPED).

[11]  Wei Chen,et al.  GreenGPU: A Holistic Approach to Energy Efficiency in GPU-CPU Heterogeneous Architectures , 2012, 2012 41st International Conference on Parallel Processing.

[12]  Kevin Skadron,et al.  Rodinia: A benchmark suite for heterogeneous computing , 2009, 2009 IEEE International Symposium on Workload Characterization (IISWC).

[13]  G. D. Peterson,et al.  Power Aware Computing on GPUs , 2012, 2012 Symposium on Application Accelerators in High Performance Computing.

[14]  Margaret Martonosi,et al.  Starchart: Hardware and software optimization using recursive partitioning regression trees , 2013, Proceedings of the 22nd International Conference on Parallel Architectures and Compilation Techniques.

[15]  Martin Schulz,et al.  Adaptive Configuration Selection for Power-Constrained Heterogeneous Systems , 2014, 2014 43rd International Conference on Parallel Processing.

[16]  Satoshi Matsuoka,et al.  Statistical power modeling of GPU kernels using performance counters , 2010, International Conference on Green Computing.

[17]  Shuaiwen Song,et al.  A Simplified and Accurate Model of Power-Performance Efficiency on Emergent GPU Architectures , 2013, 2013 IEEE 27th International Symposium on Parallel and Distributed Processing.

[18]  Karthikeyan Sankaralingam,et al.  Dark Silicon and the End of Multicore Scaling , 2012, IEEE Micro.

[19]  Sunita Chandrasekaran,et al.  Statistical modeling of power/energy of scientific kernels on a multi-GPU system , 2013, 2013 International Green Computing Conference Proceedings.

[20]  Margaret Martonosi,et al.  Stargazer: Automated regression-based GPU design space exploration , 2012, 2012 IEEE International Symposium on Performance Analysis of Systems & Software.

[21]  Petru Eles,et al.  General purpose computing on low-power embedded GPUs: Has it come of age? , 2013, 2013 International Conference on Embedded Computer Systems: Architectures, Modeling, and Simulation (SAMOS).

[22]  Frank Mueller,et al.  A Power-Aware Cost Model for HPC Procurement , 2016, 2016 IEEE International Parallel and Distributed Processing Symposium Workshops (IPDPSW).

[23]  Collin McCurdy,et al.  The Scalable Heterogeneous Computing (SHOC) benchmark suite , 2010, GPGPU-3.

[24]  Martin Schulz,et al.  Exploring hardware overprovisioning in power-constrained, high performance computing , 2013, ICS '13.

[25]  Lieven Eeckhout,et al.  Computer Architecture Performance Evaluation Methods , 2010, Computer Architecture Performance Evaluation Methods.

[26]  Henry Hoffmann,et al.  Maximizing Performance Under a Power Cap: A Comparison of Hardware, Software, and Hybrid Techniques , 2016, ASPLOS.

[27]  Hiroshi Sasaki,et al.  Power and Performance Characterization and Modeling of GPU-Accelerated Systems , 2014, 2014 IEEE 28th International Parallel and Distributed Processing Symposium.