Real-time scheduling of parallel tasks in the Linux Kernel

Bibliographic Details
Main Author: Fonseca, José
Publication Date: 2012
Other Authors: Nogueira, Luís, Maia, Cláudio, Pinho, Luís Miguel
Format: Report
Language: eng
Source: Repositórios Científicos de Acesso Aberto de Portugal (RCAAP)
Download full: http://hdl.handle.net/10400.22/3693
Summary: This paper proposes a global multiprocessor scheduling algorithm for the Linux kernel that combines the global EDF scheduler with a priority-aware work-stealing load balancing scheme, enabling parallel real-time tasks to be executed on more than one processor at a given time instant. We state that some priority inversion may actually be acceptable, provided it helps reduce contention, communication, synchronisation and coordination between parallel threads, while still guaranteeing the expected system’s predictability. Experimental results demonstrate the low scheduling overhead of the proposed approach comparatively to an existing real-time deadline-oriented scheduling class for the Linux kernel.
id RCAP_3b7e68fc98f86c6edbc527804dcdf8a7
oai_identifier_str oai:recipp.ipp.pt:10400.22/3693
network_acronym_str RCAP
network_name_str Repositórios Científicos de Acesso Aberto de Portugal (RCAAP)
repository_id_str https://opendoar.ac.uk/repository/7160
spelling Real-time scheduling of parallel tasks in the Linux KernelReal-time schedulingTask-level parallelismWork-stealingLinuxThis paper proposes a global multiprocessor scheduling algorithm for the Linux kernel that combines the global EDF scheduler with a priority-aware work-stealing load balancing scheme, enabling parallel real-time tasks to be executed on more than one processor at a given time instant. We state that some priority inversion may actually be acceptable, provided it helps reduce contention, communication, synchronisation and coordination between parallel threads, while still guaranteeing the expected system’s predictability. Experimental results demonstrate the low scheduling overhead of the proposed approach comparatively to an existing real-time deadline-oriented scheduling class for the Linux kernel.IPP Hurray! Research GroupREPOSITÓRIO P.PORTOFonseca, JoséNogueira, LuísMaia, CláudioPinho, Luís Miguel2014-02-04T16:02:25Z20122012-01-01T00:00:00Zinfo:eu-repo/semantics/publishedVersioninfo:eu-repo/semantics/reportapplication/pdfhttp://hdl.handle.net/10400.22/3693enginfo:eu-repo/semantics/openAccessreponame:Repositórios Científicos de Acesso Aberto de Portugal (RCAAP)instname:FCCN, serviços digitais da FCT – Fundação para a Ciência e a Tecnologiainstacron:RCAAP2025-04-02T02:57:56Zoai:recipp.ipp.pt:10400.22/3693Portal AgregadorONGhttps://www.rcaap.pt/oai/openaireinfo@rcaap.ptopendoar:https://opendoar.ac.uk/repository/71602025-05-29T00:30:59.434186Repositórios Científicos de Acesso Aberto de Portugal (RCAAP) - FCCN, serviços digitais da FCT – Fundação para a Ciência e a Tecnologiafalse
dc.title.none.fl_str_mv Real-time scheduling of parallel tasks in the Linux Kernel
title Real-time scheduling of parallel tasks in the Linux Kernel
spellingShingle Real-time scheduling of parallel tasks in the Linux Kernel
Fonseca, José
Real-time scheduling
Task-level parallelism
Work-stealing
Linux
title_short Real-time scheduling of parallel tasks in the Linux Kernel
title_full Real-time scheduling of parallel tasks in the Linux Kernel
title_fullStr Real-time scheduling of parallel tasks in the Linux Kernel
title_full_unstemmed Real-time scheduling of parallel tasks in the Linux Kernel
title_sort Real-time scheduling of parallel tasks in the Linux Kernel
author Fonseca, José
author_facet Fonseca, José
Nogueira, Luís
Maia, Cláudio
Pinho, Luís Miguel
author_role author
author2 Nogueira, Luís
Maia, Cláudio
Pinho, Luís Miguel
author2_role author
author
author
dc.contributor.none.fl_str_mv REPOSITÓRIO P.PORTO
dc.contributor.author.fl_str_mv Fonseca, José
Nogueira, Luís
Maia, Cláudio
Pinho, Luís Miguel
dc.subject.por.fl_str_mv Real-time scheduling
Task-level parallelism
Work-stealing
Linux
topic Real-time scheduling
Task-level parallelism
Work-stealing
Linux
description This paper proposes a global multiprocessor scheduling algorithm for the Linux kernel that combines the global EDF scheduler with a priority-aware work-stealing load balancing scheme, enabling parallel real-time tasks to be executed on more than one processor at a given time instant. We state that some priority inversion may actually be acceptable, provided it helps reduce contention, communication, synchronisation and coordination between parallel threads, while still guaranteeing the expected system’s predictability. Experimental results demonstrate the low scheduling overhead of the proposed approach comparatively to an existing real-time deadline-oriented scheduling class for the Linux kernel.
publishDate 2012
dc.date.none.fl_str_mv 2012
2012-01-01T00:00:00Z
2014-02-04T16:02:25Z
dc.type.status.fl_str_mv info:eu-repo/semantics/publishedVersion
dc.type.driver.fl_str_mv info:eu-repo/semantics/report
format report
status_str publishedVersion
dc.identifier.uri.fl_str_mv http://hdl.handle.net/10400.22/3693
url http://hdl.handle.net/10400.22/3693
dc.language.iso.fl_str_mv eng
language eng
dc.rights.driver.fl_str_mv info:eu-repo/semantics/openAccess
eu_rights_str_mv openAccess
dc.format.none.fl_str_mv application/pdf
dc.publisher.none.fl_str_mv IPP Hurray! Research Group
publisher.none.fl_str_mv IPP Hurray! Research Group
dc.source.none.fl_str_mv reponame:Repositórios Científicos de Acesso Aberto de Portugal (RCAAP)
instname:FCCN, serviços digitais da FCT – Fundação para a Ciência e a Tecnologia
instacron:RCAAP
instname_str FCCN, serviços digitais da FCT – Fundação para a Ciência e a Tecnologia
instacron_str RCAAP
institution RCAAP
reponame_str Repositórios Científicos de Acesso Aberto de Portugal (RCAAP)
collection Repositórios Científicos de Acesso Aberto de Portugal (RCAAP)
repository.name.fl_str_mv Repositórios Científicos de Acesso Aberto de Portugal (RCAAP) - FCCN, serviços digitais da FCT – Fundação para a Ciência e a Tecnologia
repository.mail.fl_str_mv info@rcaap.pt
_version_ 1833600571204960256

Similar Items