A Study of Hardware Performance Counters Selection for Cross Architectural GPU Power Modeling

In the exascale race where huge corporations are spending billions of dollars on designing highly efficient heterogeneous supercomputers, the real need to reduce power envelopes forces current technologies to face crucial challenges as well as it demands the scientific community to evaluate and opti...

Descripción completa

Guardado en:
Detalles Bibliográficos
Autores principales: Pi Puig, Martín, De Giusti, Laura Cristina, Naiouf, Marcelo, De Giusti, Armando Eduardo
Formato: Objeto de conferencia
Lenguaje:Inglés
Publicado: 2019
Materias:
Ciencias Informáticas
GPU
Architectures
Performance Counters
Power Consumption
Correlation
Prediction
Acceso en línea:http://sedici.unlp.edu.ar/handle/10915/90904
Aporte de:
SEDICI (UNLP) de Universidad Nacional de La Plata
id I19-R120-10915-90904
record_format dspace
institution Universidad Nacional de La Plata
institution_str I-19
repository_str R-120
collection SEDICI (UNLP)
language Inglés
topic Ciencias Informáticas
GPU
Architectures
Performance Counters
Power Consumption
Correlation
Prediction
spellingShingle Ciencias Informáticas
GPU
Architectures
Performance Counters
Power Consumption
Correlation
Prediction
Pi Puig, Martín
De Giusti, Laura Cristina
Naiouf, Marcelo
De Giusti, Armando Eduardo
A Study of Hardware Performance Counters Selection for Cross Architectural GPU Power Modeling
topic_facet Ciencias Informáticas
GPU
Architectures
Performance Counters
Power Consumption
Correlation
Prediction
description In the exascale race where huge corporations are spending billions of dollars on designing highly efficient heterogeneous supercomputers, the real need to reduce power envelopes forces current technologies to face crucial challenges as well as it demands the scientific community to evaluate and optimize the performance-power ratio. While energy consumption continues to climb up, the viability of these massive systems becomes a growing concern. In this context, the relevance of specific power-related research works turns into a priority. So we here develop an exhaustive step-by-step process for selecting a comprehensive set of hardware performance counters to serve as an input in an eventual GPU cross-architectural power consumption model. Our experiments show a high power-performance correlation between shared GPU events. Also, we present a set of events that delivers exclusive performance information in order to predict accurately GPU power fluctuations.
format Objeto de conferencia
Objeto de conferencia
author Pi Puig, Martín
De Giusti, Laura Cristina
Naiouf, Marcelo
De Giusti, Armando Eduardo
author_facet Pi Puig, Martín
De Giusti, Laura Cristina
Naiouf, Marcelo
De Giusti, Armando Eduardo
author_sort Pi Puig, Martín
title A Study of Hardware Performance Counters Selection for Cross Architectural GPU Power Modeling
title_short A Study of Hardware Performance Counters Selection for Cross Architectural GPU Power Modeling
title_full A Study of Hardware Performance Counters Selection for Cross Architectural GPU Power Modeling
title_fullStr A Study of Hardware Performance Counters Selection for Cross Architectural GPU Power Modeling
title_full_unstemmed A Study of Hardware Performance Counters Selection for Cross Architectural GPU Power Modeling
title_sort study of hardware performance counters selection for cross architectural gpu power modeling
publishDate 2019
url http://sedici.unlp.edu.ar/handle/10915/90904
work_keys_str_mv AT pipuigmartin astudyofhardwareperformancecountersselectionforcrossarchitecturalgpupowermodeling
AT degiustilauracristina astudyofhardwareperformancecountersselectionforcrossarchitecturalgpupowermodeling
AT naioufmarcelo astudyofhardwareperformancecountersselectionforcrossarchitecturalgpupowermodeling
AT degiustiarmandoeduardo astudyofhardwareperformancecountersselectionforcrossarchitecturalgpupowermodeling
AT pipuigmartin studyofhardwareperformancecountersselectionforcrossarchitecturalgpupowermodeling
AT degiustilauracristina studyofhardwareperformancecountersselectionforcrossarchitecturalgpupowermodeling
AT naioufmarcelo studyofhardwareperformancecountersselectionforcrossarchitecturalgpupowermodeling
AT degiustiarmandoeduardo studyofhardwareperformancecountersselectionforcrossarchitecturalgpupowermodeling
bdutipo_str Repositorios
_version_ 1764820490511187970

Ejemplares similares