On the energy efficiency of graphics processing units for scientific computing

Abstract

The graphics processing unit (GPU) has emerged as a computational accelerator that dramatically reduces the time to discovery in high-end computing (HEC). However, while today's state-of-the-art GPU can easily reduce the execution time of a parallel code by many orders of magnitude, it arguably comes at the expense of significant power and energy consumption. For example, the NVIDIA GTX 280 video card is rated at 236 watts, which is as much as the rest of a compute node, thus requiring a 500-W power supply. As a consequence, the GPU has been viewed as a "nongreen" computing solution.This paper seeks to characterize, and perhaps debunk, the notion of a "power-hungry GPU" via an empirical study of the performance, power, and energy characteristics of GPUs for scientific computing. Specifically, we take an important biological code that runs in a traditional CPU environment and transform and map it to a hybrid CPU+GPU environment. The end result is that our hybrid CPU+GPU environment, hereafter referred to simply as GPU environment, delivers an energy-delay product that is multiple orders of magnitude better than a traditional CPU environment, whether unicore or multicore.

A GPU implementation of inclusion-based points-to analysis

“…In addition, our work adds another data point to the ongoing debate regarding the performance of CPU and GPU architectures and their associated programming models [10,20,22,34], and it confirms that modern GPUs can be used to accelerate a wide range of applications.…”

The Green Practitioner: A decision-making tool for green ICT

“…Likewise one could focus on more technical, ICT-related approaches such as introducing parallelization of operations using the GPU in order to make calculations more efficient [5], or applying cloud computing technology to significantly reduce hardware and software resources needed for individuals [7]. The body of work discovering and documenting these practices is continuously growing.…”

Untitled

“…The established model is able to identify important factors to the GPU power consumption, while providing accurate prediction for the runtime power from observed execution events. Huang et al [31] evaluate the performance, energy consumption and energy efficiency of commercial GPUs running scientific computing benchmarks. They demonstrate that the energy consumption of a hybrid CPU+GPU environment is significantly less than that of traditional CPU implementations.…”