TY - GEN
T1 - EPPMiner
T2 - 8th ACM International Conference on Future Energy Systems, e-Energy 2017
AU - WANG, Qiang
AU - Xu, Pengfei
AU - Zhang, Yatao
AU - CHU, Xiaowen
N1 - Publisher Copyright:
© 2017 Association for Computing Machinery.
PY - 2017/5/16
Y1 - 2017/5/16
N2 - To address the ever-increasing demand for computing capacities, more and more heterogeneous systems have been designed to use both general-purpose and special-purpose processors. On the other hand, the huge energy consumption of these heterogeneous systems raises new environmental concerns and challenges. Besides performance, energy efficiency is now another key factor to be considered by system designers and also consumers. In this paper, we present a benchmark suite EPPMiner for evaluating the performance, power, and energy of different heterogeneous systems. EPPMiner consists of 16 benchmark programs that cover a broad range of application domains, and it shows a great variety in the intensity of utilizing the processors. We have implemented a prototype of EPPMiner that supports OpenMP, CUDA, and OpenCL, and demonstrated its usage by three showcases. Firstly, we use EPPMiner to compare the power efficiency of a set of processors, including two Intel x86 CPUs, two Nvidia GPUs, and one AMD GPU. Secondly, we investigate the impact of multi-threading on the power efficiency of multi-core CPUs. At last, we use EPPMiner to illustrate the effectiveness of GPU Dynamic Voltage and Frequency Scaling (DVFS) on the power efficiency of GPGPU applications. We show that DVFS can improve the energy efficiency by 86% over the default setting on an AMD GPU.
AB - To address the ever-increasing demand for computing capacities, more and more heterogeneous systems have been designed to use both general-purpose and special-purpose processors. On the other hand, the huge energy consumption of these heterogeneous systems raises new environmental concerns and challenges. Besides performance, energy efficiency is now another key factor to be considered by system designers and also consumers. In this paper, we present a benchmark suite EPPMiner for evaluating the performance, power, and energy of different heterogeneous systems. EPPMiner consists of 16 benchmark programs that cover a broad range of application domains, and it shows a great variety in the intensity of utilizing the processors. We have implemented a prototype of EPPMiner that supports OpenMP, CUDA, and OpenCL, and demonstrated its usage by three showcases. Firstly, we use EPPMiner to compare the power efficiency of a set of processors, including two Intel x86 CPUs, two Nvidia GPUs, and one AMD GPU. Secondly, we investigate the impact of multi-threading on the power efficiency of multi-core CPUs. At last, we use EPPMiner to illustrate the effectiveness of GPU Dynamic Voltage and Frequency Scaling (DVFS) on the power efficiency of GPGPU applications. We show that DVFS can improve the energy efficiency by 86% over the default setting on an AMD GPU.
KW - Benchmarks
KW - Dynamic Voltage and Frequency Scaling
KW - Graphics Processing Units
KW - Performance Evaluation
UR - http://www.scopus.com/inward/record.url?scp=85021412796&partnerID=8YFLogxK
U2 - 10.1145/3077839.3077858
DO - 10.1145/3077839.3077858
M3 - Conference proceeding
AN - SCOPUS:85021412796
T3 - e-Energy 2017 - Proceedings of the 8th International Conference on Future Energy Systems
SP - 23
EP - 33
BT - e-Energy 2017 - Proceedings of the 8th International Conference on Future Energy Systems
PB - Association for Computing Machinery (ACM)
Y2 - 16 May 2017 through 19 May 2017
ER -
