Are GPUs Non-Green Computing Devices?/?Son las GPUs dispositivos eficientes energeticamente?

With energy consumption emerging as one of the biggest issues in the development of HPC (High Performance Computing) applications, the importance of detailed power-related research works becomes a priority. In the last years, GPU coprocessors have been increasingly used to accelerate many of these h...

Full description

Saved in:
Bibliographic Details
Published inJournal of Computer Science & Technology Vol. 18; no. 2; p. 153
Main Authors Puig, Ma, Giusti, Laura De, Naiouf, Marcelo
Format Journal Article
LanguageSpanish
Published Graduate Network of Argentine Universities with Computer Science Schools (RedUNCI) 01.10.2018
Subjects
Online AccessGet full text

Cover

Loading…
More Information
Summary:With energy consumption emerging as one of the biggest issues in the development of HPC (High Performance Computing) applications, the importance of detailed power-related research works becomes a priority. In the last years, GPU coprocessors have been increasingly used to accelerate many of these high-priced systems even though they are embedding millions of transistors on their chips delivering an immediate increase on power consumption necessities. This paper analyzes a set of applications from the Rodinia benchmark suite in terms of CPU and GPU performance and energy consumption. Specifically, it compares single-threaded and multi-threaded CPU versions with GPU implementations, and characterize the execution time, true instant power and average energy consumption to test the idea that GPUs are power-hungry computing devices. Keywords: Power, Rodinia, GPU, NVML, RAPL. Con el consumo de energia emergiendo como uno de los mayores problemas en el desarrollo de aplicaciones HPC (High Performance Computing), la importancia de trabajos especificos de investigacion en este campo se convierte en una prioridad. En los ultimos anos, los coprocesadores GPU se han utilizado frecuentemente para acelerar muchos de estos costosos sistemas, a pesar de que incorporan millones de transistores en sus chips, lo que genera un aumento considerable en los requerimientos de energia. Este articulo analiza un conjunto de aplicaciones del benchmark Rodinia en terminos de rendimiento y consumo de energia de CPU y GPU. Especificamente, se comparan las versiones secuenciales y multihilo en CPU con implementaciones GPU, caracterizando el tiempo de ejecucion, la potencia real instantanea y el consumo promedio de energia, con el objetivo de probar la idea de que las GPU son dispositivos de baja eficiencia energetica. Palabras claves: Potencia, Rodinia, GPU, NVML, RAPL.
ISSN:1666-6046
DOI:10.24215/16666038.18.e17