• JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
 
  Bookmark and Share
 
 
Doctoral Thesis
DOI
https://doi.org/10.11606/T.55.2015.tde-10092015-144229
Document
Author
Full name
Adriana Molina Centurion
E-mail
Institute/School/College
Knowledge Area
Date of Defense
Published
São Carlos, 2015
Supervisor
Committee
Santana, Regina Helena Carlucci (President)
Bruschi, Sarita Mazzini
Sato, Liria Matsumoto
Teixeira, Fabiano Costa
Travieso, Gonzalo
Title in Portuguese
Impacto das rajadas no desempenho de serviços executados em ambientes em nuvens
Keywords in Portuguese
Avaliação de desempenho
Cargas de trabalho
Computação em nuvem
Fenômeno de rajadas
Abstract in Portuguese
Esta tese apresenta a caracterização de desempenho dos serviços executados em um ambiente em nuvem, quando são consideradas rajadas de diferentes origens, intensidades e variabilidades nas cargas de trabalho. Os resultados mostram que a presença de rajadas no processo de chegada das requisições e/ou nas demandas de serviço, ocasiona uma considerável degradação no desempenho dos serviços e, portanto, devem ser consideradas nos modelos de cargas de trabalhos e nas atividades voltadas para avaliação de desempenho em computação em nuvem. Considerando-se a grande influência das rajadas, é proposta e validada uma metodologia que permite monitorar uma carga de trabalho e determinar a ocorrência de rajadas tanto nas taxas de chegadas de requisições quanto nas demandas de serviços. A metodologia utilizada na condução deste trabalho consta de diferentes modelos de cargas de trabalho com rajadas de diferentes variabilidades e intensidades, desenvolvidos e integrados à arquitetura CloudSim-BEQoS proposta nesta tese. Utilizando-se essa arquitetura é possível executar um conjunto de experimentos que possibilitam a obtenção dos resultados que caracterizam o desempenho dos serviços quando são criadas condições de rajadas nas cargas de trabalho submetidas à nuvem.
Title in English
The impact of bursts in the performance of services executed in cloud environments
Keywords in English
Cloud computing
Performance evaluation
Phenomenon of bursts
Workload
Abstract in English
This thesis presents the performance characterization of the services executed in a cloud environment, when bursts are considered from different sources, intensity and variability in the workload. The results show that the presence of bursts in the arrival process of requests and/or in service demands, causes a significant degradation in the performance of services and therefore should belong to the models of workloads and in the activities considered for performance evaluation in cloud computing. Considering the great influence of bursts, a methodology to monitor a workload and predict the occurrence of bursts in both the rates of request arrivals and the service demands is proposed and validated. The methodology used in the conduction of this work consists in different types of workloads with bursts of different variability and intensity, developed and integrated into the CloudSim-BEQoS architecture proposed in this thesis. By using this architecture it is possible to execute a set of experiments that enable the achievement of the results that characterize the performance of services when bursts conditions are created in the workload submitted to the cloud.
 
WARNING - Viewing this document is conditioned on your acceptance of the following terms of use:
This document is only for private use for research and teaching activities. Reproduction for commercial use is forbidden. This rights cover the whole data about this document as well as its contents. Any uses or copies of this document in whole or in part must include the author's name.
Publishing Date
2015-09-10
 
WARNING: Learn what derived works are clicking here.
All rights of the thesis/dissertation are from the authors
CeTI-SC/STI
Digital Library of Theses and Dissertations of USP. Copyright © 2001-2024. All rights reserved.