Towards to virtual infrastructure allocation on multiple IaaS providers with survivability and reliability requirements

Bibliographic Details
Main Author: Raugust A.S.*
Publication Date: 2019
Other Authors: Loch W.J.*, Rodrigo De Souza F., Pillon, Mauricio Aronne, Miers, Charles Christian, Koslovski, Guilherme Piegas
Format: Article
Language: eng
Source: Repositório Institucional da Udesc
dARK ID: ark:/33523/0013000006f50
Download full: https://repositorio.udesc.br/handle/UDESC/5762
Summary: Copyright © 2019 Inderscience Enterprises Ltd.The diversity of services, prices, and geographical footprints has turned the clouds into a complex and heterogeneous environment. Moreover, the survivability and reliability aspects are often disregarded by tenants, eventually resulting in heavy losses due to unavailability of services that are hosted on Virtual Infrastructures (VIs). We present an alternative to improve Vis’ survivability and reliability, which considers the use of replicas and the spreading of virtual resources atop providers, regions, and zones. We formulate the VI allocation with survivability and reliability requirements as a mixed integer program, and three strategies to solve the formulation are proposed. First, the binary constraints are relaxed to obtain a Linear Program (LP), and the LP solution is given as input for the simulated annealing technique. Then, two GPU-accelerated algorithms are proposed to speed up the allocation of large-scale scenarios. Simulation with different reliability requests indicate an increasing in survivability without inflating costs.
id UDESC-2_49a43c680abb1b539e95e1e82dfe070b
oai_identifier_str oai:repositorio.udesc.br:UDESC/5762
network_acronym_str UDESC-2
network_name_str Repositório Institucional da Udesc
repository_id_str 6391
spelling Towards to virtual infrastructure allocation on multiple IaaS providers with survivability and reliability requirementsCopyright © 2019 Inderscience Enterprises Ltd.The diversity of services, prices, and geographical footprints has turned the clouds into a complex and heterogeneous environment. Moreover, the survivability and reliability aspects are often disregarded by tenants, eventually resulting in heavy losses due to unavailability of services that are hosted on Virtual Infrastructures (VIs). We present an alternative to improve Vis’ survivability and reliability, which considers the use of replicas and the spreading of virtual resources atop providers, regions, and zones. We formulate the VI allocation with survivability and reliability requirements as a mixed integer program, and three strategies to solve the formulation are proposed. First, the binary constraints are relaxed to obtain a Linear Program (LP), and the LP solution is given as input for the simulated annealing technique. Then, two GPU-accelerated algorithms are proposed to speed up the allocation of large-scale scenarios. Simulation with different reliability requests indicate an increasing in survivability without inflating costs.2024-12-06T12:40:38Z2019info:eu-repo/semantics/publishedVersioninfo:eu-repo/semantics/articlep. 708 - 7231741-848810.1504/IJGUC.2019.102752https://repositorio.udesc.br/handle/UDESC/5762ark:/33523/0013000006f50International Journal of Grid and Utility Computing106Raugust A.S.*Loch W.J.*Rodrigo De Souza F.Pillon, Mauricio AronneMiers, Charles ChristianKoslovski, Guilherme Piegasengreponame:Repositório Institucional da Udescinstname:Universidade do Estado de Santa Catarina (UDESC)instacron:UDESCinfo:eu-repo/semantics/openAccess2024-12-07T20:48:43Zoai:repositorio.udesc.br:UDESC/5762Biblioteca Digital de Teses e Dissertaçõeshttps://pergamumweb.udesc.br/biblioteca/index.phpPRIhttps://repositorio-api.udesc.br/server/oai/requestri@udesc.bropendoar:63912024-12-07T20:48:43Repositório Institucional da Udesc - Universidade do Estado de Santa Catarina (UDESC)false
dc.title.none.fl_str_mv Towards to virtual infrastructure allocation on multiple IaaS providers with survivability and reliability requirements
title Towards to virtual infrastructure allocation on multiple IaaS providers with survivability and reliability requirements
spellingShingle Towards to virtual infrastructure allocation on multiple IaaS providers with survivability and reliability requirements
Raugust A.S.*
title_short Towards to virtual infrastructure allocation on multiple IaaS providers with survivability and reliability requirements
title_full Towards to virtual infrastructure allocation on multiple IaaS providers with survivability and reliability requirements
title_fullStr Towards to virtual infrastructure allocation on multiple IaaS providers with survivability and reliability requirements
title_full_unstemmed Towards to virtual infrastructure allocation on multiple IaaS providers with survivability and reliability requirements
title_sort Towards to virtual infrastructure allocation on multiple IaaS providers with survivability and reliability requirements
author Raugust A.S.*
author_facet Raugust A.S.*
Loch W.J.*
Rodrigo De Souza F.
Pillon, Mauricio Aronne
Miers, Charles Christian
Koslovski, Guilherme Piegas
author_role author
author2 Loch W.J.*
Rodrigo De Souza F.
Pillon, Mauricio Aronne
Miers, Charles Christian
Koslovski, Guilherme Piegas
author2_role author
author
author
author
author
dc.contributor.author.fl_str_mv Raugust A.S.*
Loch W.J.*
Rodrigo De Souza F.
Pillon, Mauricio Aronne
Miers, Charles Christian
Koslovski, Guilherme Piegas
description Copyright © 2019 Inderscience Enterprises Ltd.The diversity of services, prices, and geographical footprints has turned the clouds into a complex and heterogeneous environment. Moreover, the survivability and reliability aspects are often disregarded by tenants, eventually resulting in heavy losses due to unavailability of services that are hosted on Virtual Infrastructures (VIs). We present an alternative to improve Vis’ survivability and reliability, which considers the use of replicas and the spreading of virtual resources atop providers, regions, and zones. We formulate the VI allocation with survivability and reliability requirements as a mixed integer program, and three strategies to solve the formulation are proposed. First, the binary constraints are relaxed to obtain a Linear Program (LP), and the LP solution is given as input for the simulated annealing technique. Then, two GPU-accelerated algorithms are proposed to speed up the allocation of large-scale scenarios. Simulation with different reliability requests indicate an increasing in survivability without inflating costs.
publishDate 2019
dc.date.none.fl_str_mv 2019
2024-12-06T12:40:38Z
dc.type.status.fl_str_mv info:eu-repo/semantics/publishedVersion
dc.type.driver.fl_str_mv info:eu-repo/semantics/article
format article
status_str publishedVersion
dc.identifier.uri.fl_str_mv 1741-8488
10.1504/IJGUC.2019.102752
https://repositorio.udesc.br/handle/UDESC/5762
dc.identifier.dark.fl_str_mv ark:/33523/0013000006f50
identifier_str_mv 1741-8488
10.1504/IJGUC.2019.102752
ark:/33523/0013000006f50
url https://repositorio.udesc.br/handle/UDESC/5762
dc.language.iso.fl_str_mv eng
language eng
dc.relation.none.fl_str_mv International Journal of Grid and Utility Computing
10
6
dc.rights.driver.fl_str_mv info:eu-repo/semantics/openAccess
eu_rights_str_mv openAccess
dc.format.none.fl_str_mv p. 708 - 723
dc.source.none.fl_str_mv reponame:Repositório Institucional da Udesc
instname:Universidade do Estado de Santa Catarina (UDESC)
instacron:UDESC
instname_str Universidade do Estado de Santa Catarina (UDESC)
instacron_str UDESC
institution UDESC
reponame_str Repositório Institucional da Udesc
collection Repositório Institucional da Udesc
repository.name.fl_str_mv Repositório Institucional da Udesc - Universidade do Estado de Santa Catarina (UDESC)
repository.mail.fl_str_mv ri@udesc.br
_version_ 1842258093221609472