Line: Evaluating Software Applications in Unreliable Environments

Título traducido de la contribución: linea: Evaluación de aplicaciones de software en entornos poco fiables

Juan F. Perez, Giuliano Casale

Resultado de la investigación: Contribución a RevistaArtículo

3 Citas (Scopus)

Resumen

La computación en nube ha allanado el camino para el despliegue flexible de aplicaciones de software. Esta flexibilidad ofrece a los proveedores de servicios una serie de opciones para adaptar sus despliegues a las cargas de trabajo de los clientes observadas y previstas, sin incurrir en grandes costes de capital. Sin embargo, los despliegues en la nube plantean nuevos desafíos en cuanto a la fiabilidad y el rendimiento de las aplicaciones. Algunos ejemplos son la gestión de la fiabilidad de las implementaciones que utilizan instancias puntuales o la gestión de la variabilidad de rendimiento causada por múltiples inquilinos en un entorno virtualizado. En este artículo, presentamos Line, una herramienta para el análisis del rendimiento y la fiabilidad de las aplicaciones de software. Line resuelve modelos de redes de colas en capas (LQN), una clase popular de modelos estocásticos en la ingeniería del rendimiento del software, mediante la configuración y resolución de un sistema asociado de ecuaciones diferenciales ordinarias. Un diferenciador clave de Line en comparación con los solucionadores existentes para LQNs es que Line incorpora un modelo del entorno en el que opera la aplicación. Esto permite el modelado de problemas de confiabilidad y rendimiento tales como fallas de recursos, averías y reparaciones del servidor, tiempos de arranque lentos, interferencia de recursos debido a la multitenencia, entre otros. Este documento describe la herramienta Line, su soporte para el modelado de rendimiento y fiabilidad, e ilustra su potencial comparando las predicciones de la línea con los datos obtenidos de una implementación en nube. También ilustramos la aplicabilidad de Line con un estudio de caso sobre el aprovisionamiento de recursos consciente de la fiabilidad.
Idioma originalEnglish (US)
Número de artículo7843645
Páginas (desde-hasta)837-853
Número de páginas17
PublicaciónIEEE Transactions on Reliability
Volumen66
N.º3
DOI
EstadoPublished - sep 1 2017

Huella dactilar

Application programs
Queueing networks
Stochastic models
Cloud computing
Reliability analysis
Ordinary differential equations
Repair
Servers
Costs

All Science Journal Classification (ASJC) codes

  • Safety, Risk, Reliability and Quality
  • Electrical and Electronic Engineering

Citar esto

@article{6feff9b819fe4bd2872cf9e2c7df6e34,
title = "Line: Evaluating Software Applications in Unreliable Environments",
abstract = "Cloud computing has paved the way to the flexible deployment of software applications. This flexibility offers service providers a number of options to tailor their deployments to the observed and foreseen customer workloads, without incurring in large capital costs. However, cloud deployments pose novel challenges regarding application reliability and performance. Examples include managing the reliability of deployments that make use of spot instances, or coping with the performance variability caused by multiple tenants in a virtualized environment. In this paper, we introduce Line, a tool for performance and reliability analysis of software applications. Line solves layered queueing network (LQN) models, a popular class of stochastic models in software performance engineering, by setting up and solving an associated system of ordinary differential equations. A key differentiator of Line compared to existing solvers for LQNs is that Line incorporates a model of the environment the application operates in. This enables the modeling of reliability and performance issues such as resource failures, server breakdowns and repairs, slow start-up times, resource interference due to multitenancy, among others. This paper describes the Line tool, its support for performance and reliability modeling, and illustrates its potential by comparing Line predictions against data obtained from a cloud deployment. We also illustrate the applicability of Line with a case study on reliability-aware resource provisioning.",
author = "Perez, {Juan F.} and Giuliano Casale",
year = "2017",
month = "9",
day = "1",
doi = "10.1109/TR.2017.2655505",
language = "English (US)",
volume = "66",
pages = "837--853",
journal = "IEEE Transactions on Reliability",
issn = "0018-9529",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
number = "3",

}

Line : Evaluating Software Applications in Unreliable Environments. / Perez, Juan F.; Casale, Giuliano.

En: IEEE Transactions on Reliability, Vol. 66, N.º 3, 7843645, 01.09.2017, p. 837-853.

Resultado de la investigación: Contribución a RevistaArtículo

TY - JOUR

T1 - Line

T2 - Evaluating Software Applications in Unreliable Environments

AU - Perez, Juan F.

AU - Casale, Giuliano

PY - 2017/9/1

Y1 - 2017/9/1

N2 - Cloud computing has paved the way to the flexible deployment of software applications. This flexibility offers service providers a number of options to tailor their deployments to the observed and foreseen customer workloads, without incurring in large capital costs. However, cloud deployments pose novel challenges regarding application reliability and performance. Examples include managing the reliability of deployments that make use of spot instances, or coping with the performance variability caused by multiple tenants in a virtualized environment. In this paper, we introduce Line, a tool for performance and reliability analysis of software applications. Line solves layered queueing network (LQN) models, a popular class of stochastic models in software performance engineering, by setting up and solving an associated system of ordinary differential equations. A key differentiator of Line compared to existing solvers for LQNs is that Line incorporates a model of the environment the application operates in. This enables the modeling of reliability and performance issues such as resource failures, server breakdowns and repairs, slow start-up times, resource interference due to multitenancy, among others. This paper describes the Line tool, its support for performance and reliability modeling, and illustrates its potential by comparing Line predictions against data obtained from a cloud deployment. We also illustrate the applicability of Line with a case study on reliability-aware resource provisioning.

AB - Cloud computing has paved the way to the flexible deployment of software applications. This flexibility offers service providers a number of options to tailor their deployments to the observed and foreseen customer workloads, without incurring in large capital costs. However, cloud deployments pose novel challenges regarding application reliability and performance. Examples include managing the reliability of deployments that make use of spot instances, or coping with the performance variability caused by multiple tenants in a virtualized environment. In this paper, we introduce Line, a tool for performance and reliability analysis of software applications. Line solves layered queueing network (LQN) models, a popular class of stochastic models in software performance engineering, by setting up and solving an associated system of ordinary differential equations. A key differentiator of Line compared to existing solvers for LQNs is that Line incorporates a model of the environment the application operates in. This enables the modeling of reliability and performance issues such as resource failures, server breakdowns and repairs, slow start-up times, resource interference due to multitenancy, among others. This paper describes the Line tool, its support for performance and reliability modeling, and illustrates its potential by comparing Line predictions against data obtained from a cloud deployment. We also illustrate the applicability of Line with a case study on reliability-aware resource provisioning.

UR - http://www.scopus.com/inward/record.url?scp=85012117664&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85012117664&partnerID=8YFLogxK

U2 - 10.1109/TR.2017.2655505

DO - 10.1109/TR.2017.2655505

M3 - Article

AN - SCOPUS:85012117664

VL - 66

SP - 837

EP - 853

JO - IEEE Transactions on Reliability

JF - IEEE Transactions on Reliability

SN - 0018-9529

IS - 3

M1 - 7843645

ER -