IEEE Transactions on Parallel and Distributed Systems · 2021 · 40 citations · 17 references
ComputingComputer EngineeringComputer ScienceEnergy ManagementMachine LearningSchedulingHardware AccelerationOptimization For Machine LearningPerformance OptimizationPower ManagementPower OptimizationWorkload SchedulingEnergy OptimizationEnergy-aware OptimizationEnergy-efficient Artificial IntelligenceEnergy-efficient Gpu ComputingGpu Performance ProfilingGradient BoostingOnline OptimizationWorkload
GPUs are widely used to accelerate the training of machine learning workloads. As modern machine learning models become increasingly larger, they require a longer time to train, leading to higher GPU energy consumption. This paper presents GPOEO, an online GPU energy optimization framework for machine learning training workloads. GPOEO dynamically determines the optimal energy configuration by employing novel techniques for online measurement, multi-objective prediction modeling, and search optimization. To characterize the target workload behavior, GPOEO utilizes GPU performance counters. To reduce the performance counter profiling overhead, it uses an analytical model to detect the training iteration change and only collects performance counter data when an iteration shift is detected. GPOEO employs multi-objective models based on gradient boosting and a local search algorithm to find a trade-off between execution time and energy consumption. We evaluate the GPOEO by applying it to 71 machine learning workloads from two AI benchmark suites running on an NVIDIA RTX3080Ti GPU. Compared with the NVIDIA default scheduling strategy, GPOEO delivers a mean energy saving of 16.2% with a modest average execution time increase of 5.1%.
17
A State-of-the-Art Survey on Deep Learning Theory and Architectures
Md Zahangir Alom, Tarek M. Taha, Chris Yakopcic et al. · Electronics · 2019 · 1.6K citations · Full text
Machine Learning in Compiler Optimization
Zheng Wang, Michael O’Boyle · Proceedings of the IEEE · 2018 · 192 citations · Full text
ThunderSVM: A Fast SVM Library on GPUs and CPUs
Zeyi Wen, Jiashuai Shi, Qinbin Li et al. · UWA Profiles and Research Repository (UWA) · 2018 · 178 citations · Full text
GPGPU Power Modeling for Multi-domain Voltage-Frequency Scaling
João Guerreiro, Aleksandar Ilić, Nuno Roma et al. · 2018 · 54 citations