Indicator-Directed Dynamic Power Management for Iterative Workloads on GPU-Accelerated Systems
Authors:
- 1 May 2020
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 559-568
Abstract
Modern high-performance and warehouse computing centers show strong interest in minimizing system power consumption while satisfying customer requested quality of service (QoS). Dynamic voltage and frequency scaling (DVFS) is effective for achieving this goal. Nevertheless, automating the process online is challenging due to the great complexity — today’s hardware components (e.g., CPUs, GPUs, memory, network, etc.) can be configured in several or dozens of frequency/voltage states for satisfying divergent system demands. Given their combination and the emergence of heterogeneity, searching the optimal configuration in the design space online can be time consuming. Existing work relies on detailed knowledge of workloads, either obtained from extensive offline profiling or intrusive code transformation for fine-grain profiling, unable to transparently support unknown workloads. This work bridges the gap and targets GPU-accelerated systems and applications.We focus on applications exhibiting an interesting feature - iterative or periodic, which is common among conventional HPC and emerging machine learning workloads. We propose an online dynamic power-performance (ODPP) management framework. ODPP runs on the host and dynamically adjusts GPU DVFS configurations to meet performance and power objectives without any code annotation or modification. Particularly, ODPP extracts the performance and power indicators for applications from easily obtained GPU resource utilization profiles in a short episode. The extraction method uses Fourier Transform and incurs minimal cost. ODPP further automatically constructs an accurate model that infers from the indicators how the application’s performance and power vary with GPU core and memory frequencies. Aided with the model, for both seen and unseen applications, ODPP can quickly determine the most appropriate DVFS configuration for their execution. We evaluate ODPP on an NVIDIA GPU using multiple exascale computing (ECP) and deep learning applications. Evaluation results show that ODPP can improve energy efficiency by over 30% under different QoS and improve performance by more than 8% under different power bounds.Keywords
This publication has 29 references indexed in Scilit:
- The CRISP performance model for dynamic voltage and frequency scaling in a GPGPU2015
- JouleGuard2015
- Going deeper with convolutions2015
- A Probabilistic Graphical Model-based Approach for Minimizing Energy Under Performance Constraints2015
- Equalizer: Dynamic Tuning of GPU Resources for Efficient Execution2014
- Run-Time Technique for Simultaneous Aging and Power Optimization in GPGPUs2014
- Power and Performance Characterization and Modeling of GPU-Accelerated Systems2014
- A measurement study of GPU DVFS on energy conservation2013
- Effects of Dynamic Voltage and Frequency Scaling on a K20 GPU2013
- An integrated GPU power and performance modelACM SIGARCH Computer Architecture News, 2010