Effects of Dynamic Voltage and Frequency Scaling on a K20 GPU

Ge, Rong; Vogt, Ryan; Majumder, Akm Jahangir Alam; Alam, Arif; Burtscher, Martin; Zong, Ziliang

doi:10.1109/icpp.2013.98

Cited by 89 publications

(66 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…However, the study of G-PU DVFS on energy conservation is still at an early stage. Existing DVFS research work on GPU energy consumption are mostly based on simulation [8,13,15]. In this paper, we aim to answer one question: can GPU DVFS save energy in practice?…”

Section: Introductionmentioning

confidence: 99%

A measurement study of GPU DVFS on energy conservation

Mei

Yung

Zhao

et al. 2013

Proceedings of the Workshop on Power-Aware Computing and Systems

View full text Add to dashboard Cite

Nowadays, GPUs are widely used to accelerate many high performance computing applications. Energy conservation of such computing systems has become an important research topic. Dynamic voltage/frequency scaling (DVFS) is proved to be an appealing method for saving energy for traditional computing centers. However, there is still a lack of firsthand study on the effectiveness of GPU DVFS. This paper presents a thorough measurement study that aims to explore how GPU DVFS affects the system energy consumption. We conduct experiments on a real GPU platform with 37 benchmark applications. Our results show that GPU voltage/frequency scaling is an effective approach to conserving energy. For example, by scaling down the GPU core voltage and frequency, we have achieved an average of 19.28% energy reduction compared with the default setting, while giving up no more than 4% of performance. For all tested GPU applications, core voltage scaling is significantly effective to reduce system energy consumption. Meanwhile the effects of scaling core frequency and memory frequency depend on the characteristics of GPU applications.

show abstract

Section: Introductionmentioning

confidence: 99%

A measurement study of GPU DVFS on energy conservation

Mei

Yung

Zhao

et al. 2013

Proceedings of the Workshop on Power-Aware Computing and Systems

View full text Add to dashboard Cite

show abstract

“…For instance, with the advent of multi-GPU compute nodes, where up to 8 dual GPU boards are hosted on a single 3 computing node, up to ≃ 75% of the power could be drained by GPUs, with CPUs accounting for only 5 − 10% of the total energy budget † . Consequently, as recent GPUs improve their support of DVFS, in many cases allowing for a fine grained frequency selection [2,11,12], various studies focused on the optimization space made available by tuning GPU clock frequencies.…”

Section: Related Workmentioning

confidence: 99%

Evaluation of DVFS techniques on modern HPC processors and accelerators for energy‐aware applications

Calore

Gabbana

Schifano

et al. 2017

Concurrency and Computation

View full text Add to dashboard Cite

SUMMARYEnergy efficiency is becoming increasingly important for computing systems, in particular for large scale HPC facilities. In this work we evaluate, from an user perspective, the use of Dynamic Voltage and Frequency Scaling (DVFS) techniques, assisted by the power and energy monitoring capabilities of modern processors in order to tune applications for energy efficiency. We run selected kernels and a full HPC application on two high-end processors widely used in the HPC context, namely an NVIDIA K80 GPU and an Intel Haswell CPU. We evaluate the available trade-offs between energy-to-solution and time-to-solution, attempting a function-by-function frequency tuning. We finally estimate the benefits obtainable running the full code on a HPC multi-GPU node, with respect to default clock frequency governors. We instrument our code to accurately monitor power consumption and execution time without the need of any additional hardware, and we enable it to change CPUs and GPUs clock frequencies while running. We analyze our results on the different architectures using a simple energy-performance model, and derive a number of energy saving strategies which can be easily adopted on recent high-end HPC systems for generic applications.

show abstract

“…MatrixMultiply serves as benchmark in this section. Unless stated otherwise, the matrix size is 12800 × 12800, which saturates the K20c GPU [8].…”

Section: Model Validation and Resultsmentioning

confidence: 99%

“…Most of the existing work in this domain attempts to understand and model the power and energy consumption of heterogeneous computing systems; research in GPU power management is still sparse. Studies in [8,11] experimentally investigate the power and energy behavior of applications on prototypes of GPU-accelerated systems. Other efforts analytically model GPU power with architecture-level instructions [9,13,19], hardware performance events [17,21], or algorithm parameters [11].…”

Section: Related Workmentioning

confidence: 99%

Performance and Energy Modeling for Cooperative Hybrid Computing

Feng

Burtscher

et al. 2014

2014 9th IEEE International Conference on Networking, Architecture, and Storage

Self Cite

View full text Add to dashboard Cite

Abstract-Accelerator-based heterogeneous systems can provide high performance and energy efficiency, both of which are key design goals in high performance computing. To fully realize the potential of heterogeneous architectures, software must optimally exploit the hosts' and accelerators' processing and power-saving capabilities. Yet, previous studies mainly focus on using hosts and accelerators to boost application performance. Power-saving features to improve the energy efficiency of parallel programs, such as Dynamic Voltage and Frequency Scaling (DVFS), remain largely unexplored.Recognizing that energy efficiency is a different objective than performance and should therefore be independently pursued, we study how to judiciously distribute computation between hosts and accelerators for energy optimization. We further explore energy-saving scheduling in combination with computation distribution for even larger gains. Moreover, we present PEACH, an analytical model for Performance and Energy Aware Cooperative Hybrid computing. With just a few system-and application-dependent parameters, PEACH accurately captures the performance and energy impact of computation distribution and energy-saving scheduling to quickly identify the optimal coupled strategy for achieving the best performance or the lowest energy consumption. PEACH thus eliminates the need for extensive profiling and measurement. Experimental results from two GPU-accelerated heterogeneous systems show that PEACH predicts the performance and energy of the studied codes with less than 3% error and successfully identifies the optimal strategy for a given objective.

show abstract

Effects of Dynamic Voltage and Frequency Scaling on a K20 GPU

Cited by 89 publications

References 14 publications

A measurement study of GPU DVFS on energy conservation

A measurement study of GPU DVFS on energy conservation

Evaluation of DVFS techniques on modern HPC processors and accelerators for energy‐aware applications

Performance and Energy Modeling for Cooperative Hybrid Computing

Contact Info

Product

Resources

About