Method to determine optimal hardware platforms in Human Centered Computing based on non functional requirements analysis

Mauricio González, Liliana González, Jaime Echeverri, Miguel Aristizábal, Germán Urrego, Ana Lucía Pérez

Resultado de la investigación: Contribución a una conferenciaArtículoInvestigación

2 Citas (Scopus)

Resumen

Human Centered Computing is a novel paradigm to process context information of human being's environment in which computers are invisible to the subject, providing tools and services depending on the context of each individual. An increasing interest is growing regarding embedded computers since they offer advantages related to portability, dedicated tasks, invisibility, amongst others. However, a plenty of hardware platforms are in the market, so it is complicated to determine which the best for a particular need. Benchmarks make those tasks easier by performing measurements in hardware by running applications and performing comparisons. Nevertheless, they are thought to meet some quite particular kinds of applications. Moreover, if some benchmark applications have to be merged, i.e. voice or images processing, there are not schemes to correctly measure hardware platforms. On the other hand, requirements such as reliability and availability are not commonly assessed as a dependant set in hardware platforms. In this work, we propose a novel method to select hardware architectures for Human Centered Computing based on benchmarking, genetic algorithms, weighted sums and statistical distances in order to consider non-functional requirements. © 2014 AISTI.
Idioma originalInglés estadounidense
DOI
EstadoPublicada - 1 ene 2014
EventoIberian Conference on Information Systems and Technologies, CISTI -
Duración: 1 ene 2014 → …

Conferencia

ConferenciaIberian Conference on Information Systems and Technologies, CISTI
Período1/01/14 → …

Huella dactilar

Hardware
Benchmarking
Image processing
Genetic algorithms
Availability

Citar esto

González, M., González, L., Echeverri, J., Aristizábal, M., Urrego, G., & Pérez, A. L. (2014). Method to determine optimal hardware platforms in Human Centered Computing based on non functional requirements analysis. Papel presentado en Iberian Conference on Information Systems and Technologies, CISTI, . https://doi.org/10.1109/CISTI.2014.6876901
González, Mauricio ; González, Liliana ; Echeverri, Jaime ; Aristizábal, Miguel ; Urrego, Germán ; Pérez, Ana Lucía. / Method to determine optimal hardware platforms in Human Centered Computing based on non functional requirements analysis. Papel presentado en Iberian Conference on Information Systems and Technologies, CISTI, .
@conference{f9064befcc3e402dac066381966858c5,
title = "Method to determine optimal hardware platforms in Human Centered Computing based on non functional requirements analysis",
abstract = "Human Centered Computing is a novel paradigm to process context information of human being's environment in which computers are invisible to the subject, providing tools and services depending on the context of each individual. An increasing interest is growing regarding embedded computers since they offer advantages related to portability, dedicated tasks, invisibility, amongst others. However, a plenty of hardware platforms are in the market, so it is complicated to determine which the best for a particular need. Benchmarks make those tasks easier by performing measurements in hardware by running applications and performing comparisons. Nevertheless, they are thought to meet some quite particular kinds of applications. Moreover, if some benchmark applications have to be merged, i.e. voice or images processing, there are not schemes to correctly measure hardware platforms. On the other hand, requirements such as reliability and availability are not commonly assessed as a dependant set in hardware platforms. In this work, we propose a novel method to select hardware architectures for Human Centered Computing based on benchmarking, genetic algorithms, weighted sums and statistical distances in order to consider non-functional requirements. {\circledC} 2014 AISTI.",
author = "Mauricio Gonz{\'a}lez and Liliana Gonz{\'a}lez and Jaime Echeverri and Miguel Aristiz{\'a}bal and Germ{\'a}n Urrego and P{\'e}rez, {Ana Luc{\'i}a}",
year = "2014",
month = "1",
day = "1",
doi = "10.1109/CISTI.2014.6876901",
language = "American English",
note = "Iberian Conference on Information Systems and Technologies, CISTI ; Conference date: 01-01-2014",

}

Method to determine optimal hardware platforms in Human Centered Computing based on non functional requirements analysis. / González, Mauricio; González, Liliana; Echeverri, Jaime; Aristizábal, Miguel; Urrego, Germán; Pérez, Ana Lucía.

2014. Papel presentado en Iberian Conference on Information Systems and Technologies, CISTI, .

Resultado de la investigación: Contribución a una conferenciaArtículoInvestigación

TY - CONF

T1 - Method to determine optimal hardware platforms in Human Centered Computing based on non functional requirements analysis

AU - González, Mauricio

AU - González, Liliana

AU - Echeverri, Jaime

AU - Aristizábal, Miguel

AU - Urrego, Germán

AU - Pérez, Ana Lucía

PY - 2014/1/1

Y1 - 2014/1/1

N2 - Human Centered Computing is a novel paradigm to process context information of human being's environment in which computers are invisible to the subject, providing tools and services depending on the context of each individual. An increasing interest is growing regarding embedded computers since they offer advantages related to portability, dedicated tasks, invisibility, amongst others. However, a plenty of hardware platforms are in the market, so it is complicated to determine which the best for a particular need. Benchmarks make those tasks easier by performing measurements in hardware by running applications and performing comparisons. Nevertheless, they are thought to meet some quite particular kinds of applications. Moreover, if some benchmark applications have to be merged, i.e. voice or images processing, there are not schemes to correctly measure hardware platforms. On the other hand, requirements such as reliability and availability are not commonly assessed as a dependant set in hardware platforms. In this work, we propose a novel method to select hardware architectures for Human Centered Computing based on benchmarking, genetic algorithms, weighted sums and statistical distances in order to consider non-functional requirements. © 2014 AISTI.

AB - Human Centered Computing is a novel paradigm to process context information of human being's environment in which computers are invisible to the subject, providing tools and services depending on the context of each individual. An increasing interest is growing regarding embedded computers since they offer advantages related to portability, dedicated tasks, invisibility, amongst others. However, a plenty of hardware platforms are in the market, so it is complicated to determine which the best for a particular need. Benchmarks make those tasks easier by performing measurements in hardware by running applications and performing comparisons. Nevertheless, they are thought to meet some quite particular kinds of applications. Moreover, if some benchmark applications have to be merged, i.e. voice or images processing, there are not schemes to correctly measure hardware platforms. On the other hand, requirements such as reliability and availability are not commonly assessed as a dependant set in hardware platforms. In this work, we propose a novel method to select hardware architectures for Human Centered Computing based on benchmarking, genetic algorithms, weighted sums and statistical distances in order to consider non-functional requirements. © 2014 AISTI.

U2 - 10.1109/CISTI.2014.6876901

DO - 10.1109/CISTI.2014.6876901

M3 - Paper

ER -

González M, González L, Echeverri J, Aristizábal M, Urrego G, Pérez AL. Method to determine optimal hardware platforms in Human Centered Computing based on non functional requirements analysis. 2014. Papel presentado en Iberian Conference on Information Systems and Technologies, CISTI, . https://doi.org/10.1109/CISTI.2014.6876901