Unified Power Modeling Design for Various Raspberry Pi Generations Analyzing Different Statistical Methods

Monitoring processor power is important to define strategies that allow reducing energy costs in computer systems. Today, processors have a large number of counters that allow monitoring system events such as CPU usage, memory, cache, and so forth. In previous works, it has been shown that parallel...

Descripción completa

Guardado en:
Detalles Bibliográficos
Autores principales: Paniego, Juan Manuel, Libutti, Leandro Ariel, Pi Puig, Martín, Chichizola, Franco, De Giusti, Laura Cristina, Naiouf, Marcelo, De Giusti, Armando Eduardo
Formato: Objeto de conferencia
Lenguaje:Español
Publicado: 2019
Materias:
Informática
Power
Raspberry Pi
Hardware counters
Modeling
Statistical models
Acceso en línea:http://sedici.unlp.edu.ar/handle/10915/136167
Aporte de:
SEDICI (UNLP) de Universidad Nacional de La Plata
id I19-R120-10915-136167
record_format dspace
spelling I19-R120-10915-1361672023-05-31T13:42:03Z http://sedici.unlp.edu.ar/handle/10915/136167 issn:1865-0929 issn:1865-0937 isbn:978-3-030-48325-8 Unified Power Modeling Design for Various Raspberry Pi Generations Analyzing Different Statistical Methods Paniego, Juan Manuel Libutti, Leandro Ariel Pi Puig, Martín Chichizola, Franco De Giusti, Laura Cristina Naiouf, Marcelo De Giusti, Armando Eduardo 2019 2019 2022-05-11T19:04:19Z es Informática Power Raspberry Pi Hardware counters Modeling Statistical models Monitoring processor power is important to define strategies that allow reducing energy costs in computer systems. Today, processors have a large number of counters that allow monitoring system events such as CPU usage, memory, cache, and so forth. In previous works, it has been shown that parallel application consumption can be predicted through these events, but only for a given SBC board architecture. In this article, we analyze the portability of a power prediction statistical model on a new generation of Raspberry boards. Our experiments focus on the optimizations using different statistical methods so as to systematically reduce the final estimation error in the architectures analyzed. The final models yield an average error between 2.24% and 4.45%, increasing computational cost as the prediction error decreases. Trabajo publicado en Pesado, P., Arroyo, M. (eds.). Computer Science – CACIC 2019. Communications in Computer and Information Science (CCIS), vol. 1184. Springer, Cham. Instituto de Investigación en Informática Comisión de Investigaciones Científicas de la provincia de Buenos Aires Consejo Nacional de Investigaciones Científicas y Técnicas Objeto de conferencia Objeto de conferencia http://creativecommons.org/licenses/by-nc-sa/4.0/ Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International (CC BY-NC-SA 4.0) application/pdf 53-65
institution Universidad Nacional de La Plata
institution_str I-19
repository_str R-120
collection SEDICI (UNLP)
language Español
topic Informática
Power
Raspberry Pi
Hardware counters
Modeling
Statistical models
spellingShingle Informática
Power
Raspberry Pi
Hardware counters
Modeling
Statistical models
Paniego, Juan Manuel
Libutti, Leandro Ariel
Pi Puig, Martín
Chichizola, Franco
De Giusti, Laura Cristina
Naiouf, Marcelo
De Giusti, Armando Eduardo
Unified Power Modeling Design for Various Raspberry Pi Generations Analyzing Different Statistical Methods
topic_facet Informática
Power
Raspberry Pi
Hardware counters
Modeling
Statistical models
description Monitoring processor power is important to define strategies that allow reducing energy costs in computer systems. Today, processors have a large number of counters that allow monitoring system events such as CPU usage, memory, cache, and so forth. In previous works, it has been shown that parallel application consumption can be predicted through these events, but only for a given SBC board architecture. In this article, we analyze the portability of a power prediction statistical model on a new generation of Raspberry boards. Our experiments focus on the optimizations using different statistical methods so as to systematically reduce the final estimation error in the architectures analyzed. The final models yield an average error between 2.24% and 4.45%, increasing computational cost as the prediction error decreases.
format Objeto de conferencia
Objeto de conferencia
author Paniego, Juan Manuel
Libutti, Leandro Ariel
Pi Puig, Martín
Chichizola, Franco
De Giusti, Laura Cristina
Naiouf, Marcelo
De Giusti, Armando Eduardo
author_facet Paniego, Juan Manuel
Libutti, Leandro Ariel
Pi Puig, Martín
Chichizola, Franco
De Giusti, Laura Cristina
Naiouf, Marcelo
De Giusti, Armando Eduardo
author_sort Paniego, Juan Manuel
title Unified Power Modeling Design for Various Raspberry Pi Generations Analyzing Different Statistical Methods
title_short Unified Power Modeling Design for Various Raspberry Pi Generations Analyzing Different Statistical Methods
title_full Unified Power Modeling Design for Various Raspberry Pi Generations Analyzing Different Statistical Methods
title_fullStr Unified Power Modeling Design for Various Raspberry Pi Generations Analyzing Different Statistical Methods
title_full_unstemmed Unified Power Modeling Design for Various Raspberry Pi Generations Analyzing Different Statistical Methods
title_sort unified power modeling design for various raspberry pi generations analyzing different statistical methods
publishDate 2019
url http://sedici.unlp.edu.ar/handle/10915/136167
work_keys_str_mv AT paniegojuanmanuel unifiedpowermodelingdesignforvariousraspberrypigenerationsanalyzingdifferentstatisticalmethods
AT libuttileandroariel unifiedpowermodelingdesignforvariousraspberrypigenerationsanalyzingdifferentstatisticalmethods
AT pipuigmartin unifiedpowermodelingdesignforvariousraspberrypigenerationsanalyzingdifferentstatisticalmethods
AT chichizolafranco unifiedpowermodelingdesignforvariousraspberrypigenerationsanalyzingdifferentstatisticalmethods
AT degiustilauracristina unifiedpowermodelingdesignforvariousraspberrypigenerationsanalyzingdifferentstatisticalmethods
AT naioufmarcelo unifiedpowermodelingdesignforvariousraspberrypigenerationsanalyzingdifferentstatisticalmethods
AT degiustiarmandoeduardo unifiedpowermodelingdesignforvariousraspberrypigenerationsanalyzingdifferentstatisticalmethods
_version_ 1767633948616163328

Ejemplares similares