Publication:
Source code optimizations for low power consumption on microprocessor-based systems

dc.contributor.advisor Santiago-Santiago, Nayda G.
dc.contributor.author Ortiz-López, David A.
dc.contributor.college College of Engineering en_US
dc.contributor.committee Ducoudray, Gladys O.
dc.contributor.committee Sepúlveda, Nelson
dc.contributor.department Department of Electrical and Computer Engineering en_US
dc.contributor.representative Walker, Uroyoán R.
dc.date.accessioned 2019-05-15T17:59:33Z
dc.date.available 2019-05-15T17:59:33Z
dc.date.issued 2007
dc.description.abstract Power consumption is an important constraint in the design of battery-operated embedded systems. Minimizing power dissipation may be handled in terms of hardware or software optimizations. Source code-level optimization techniques have been used as an alternative to achieve low power consumption when programming embedded systems, however these techniques should be analyzed with statistical sound methods in order to reach strong conclusions about their actual impact on the power consumption. In this work, source code optimizations are applied on a set of representative benchmarks for embedded processors (MiBench), to analyze whether the techniques have or not an effect on the power dissipation of a set of microprocessor based platforms. Design of experiments techniques (DOE) and analysis of variance (ANOVA) are used to achieve statistical sound conclusions. Results showed that not all optimizations have an effect on power consumption, moreover some techniques depend on the target platform where they are run. en_US
dc.description.abstract El consumo de potencia es un factor de gran importancia en sistemas electrónicos basados en microprocesadores. Este problema puede ser resuelto con optimizaciones en términos de hardware y software. Diversas técnicas de optimización de software han sido utilizadas como alternativa para lograr obtener bajo consumo de potencia al diseñar este tipo de sistemas electrónicos, sin embargo estas técnicas deben ser analizadas con métodos estadísticos para obtener conclusiones sólidas sobre su efecto real. En este trabajo, tres técnicas de optimización aplicadas en lenguage de alto nivel son implementadas en un conjunto de benchmarks (MiBench) que representan aplicaciones típicas de sistemas basados en microprocesadores, con el objetivo de analizar si las técnicas de optimización tienen o no efecto en el consumo de potencia de distintas plataformas. Para realizar este análisis, métodos de diseño de experimentos (DOE) y análisis de varianza (ANOVA) fueron usados para obtener conclusiones basadas en hechos estadísticos. Los resultados mostraron que no todas las optimizaciones tienen efecto en el consumo de potencia, adem ́as algunas técnicas dependen de la plataforma en la cual son ejecutadas. en_US
dc.description.graduationYear 2007 en_US
dc.description.sponsorship WIMS Center at the University of Michigan, Ann Arbor, for the sponsorship of this research. This work has been supported by the Engineering Research Centers Program of the National Science Foundation under Award Number ERC-9986866. en_US
dc.identifier.uri https://hdl.handle.net/20.500.11801/2369
dc.language.iso English en_US
dc.rights.holder (c) 2007 David Andrés Ortiz-López en_US
dc.rights.license All rights reserved en_US
dc.title Source code optimizations for low power consumption on microprocessor-based systems en_US
dc.type Thesis en_US
dspace.entity.type Publication
thesis.degree.discipline Electrical Engineering en_US
thesis.degree.level M.S. en_US
Files
Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
INEL_OrtizLopezD_2007.pdf
Size:
1.22 MB
Format:
Adobe Portable Document Format
Description: