PerfBound: Conserving Energy with Bounded Overheads in On/Off-Based HPC Interconnects

Energy and power are key challenges in high-performance computing. System energy efficiency must be significantly improved, and this requires greater efficiency in all subcomponents. An important target of optimization is the interconnect, since network links are always on, consuming power even duri...

Descripción completa

Detalles Bibliográficos
Autores: Saravanan, Karthikeyan P., Carpenter, Paul Matthew
Tipo de recurso: artículo
Fecha de publicación:2018
País:España
Institución:Universitat Politècnica de Catalunya (UPC)
Repositorio:UPCommons. Portal del coneixement obert de la UPC
Idioma:inglés
OAI Identifier:oai:upcommons.upc.edu:2117/125138
Acceso en línea:https://hdl.handle.net/2117/125138
https://dx.doi.org/10.1109/TC.2018.2790394
Access Level:acceso abierto
Palabra clave:High performance computing
Energy Efficient Interconnects
Energy Efficient Ethernet
Fast-Wake
Deep-Sleep
Supercomputadors
Àrees temàtiques de la UPC::Informàtica
id ES_95def99ad345cb399caa36f7c19ca615
oai_identifier_str oai:upcommons.upc.edu:2117/125138
network_acronym_str ES
network_name_str España
repository_id_str
spelling PerfBound: Conserving Energy with Bounded Overheads in On/Off-Based HPC InterconnectsSaravanan, Karthikeyan P.Carpenter, Paul MatthewHigh performance computingEnergy Efficient InterconnectsEnergy Efficient EthernetFast-WakeDeep-SleepSupercomputadorsÀrees temàtiques de la UPC::InformàticaEnergy and power are key challenges in high-performance computing. System energy efficiency must be significantly improved, and this requires greater efficiency in all subcomponents. An important target of optimization is the interconnect, since network links are always on, consuming power even during idle periods. A large number of HPC machines have a primary interconnect based on Ethernet (about 40 percent of TOP500 machines), which, since 2010, has included support for saving power via Energy Efficient Ethernet (EEE). Nevertheless, it is unlikely that HPC interconnects would use these energy saving modes unless the performance overhead is known and small. This paper presents PerfBound, a self-contained technique to manage on/off-based networks such as EEE, minimizing interconnect link energy consumption subject to a bound on the performance degradation. PerfBound does not require changes to the applications and it uses only local information already available at switches and NICs without introducing additional communication messages, and is also compatible with multi-hop networks. PerfBound is evaluated using traces from a production supercomputer. For twelve out of fourteen applications, PerfBound has high energy savings, up to 70 percent for only 1 percent performance degradation. This paper also presents DynamicFastwake, which extends PerfBound to exploit multiple low-power states. DynamicFastwake achieves an energy-delay product 10 percent lower than the original PerfBound techniqueThis research was supported by European Union’s 7th Framework Programme [FP7/2007-2013] under the Mont-Blanc-3 (FP7-ICT-671697) and EUROSERVER (FP7-ICT-610456) projects, the Ministry of Economy and Competitiveness of Spain (TIN2012-34557 and TIN2015-65316), Generalitat de Catalunya (FI-AGAUR 2012 FI B 00644, 2014-SGR-1051 and 2014-SGR-1272), the European Union’s Horizon2020 research and innovation programme under the HiPEAC-3 Network of Excellence (ICT-287759), and the Severo Ochoa Program (SEV-2011-00067) of the Spanish Government.Peer ReviewedIEEE20182018-07-0120182018-11-27journal articlehttp://purl.org/coar/resource_type/c_6501AMhttp://purl.org/coar/version/c_ab4af688f83e57aainfo:eu-repo/semantics/articleapplication/pdfhttps://hdl.handle.net/2117/125138https://dx.doi.org/10.1109/TC.2018.2790394reponame:UPCommons. Portal del coneixement obert de la UPCinstname:Universitat Politècnica de Catalunya (UPC)InglésengEuropean Commission http://dx.doi.org/10.13039/100011102 Seventh Framework Programme 287759 High Performance and Embedded Architecture and Compilationopen accesshttp://purl.org/coar/access_right/c_abf2info:eu-repo/semantics/openAccessoai:upcommons.upc.edu:2117/1251382026-05-27T15:37:01Z
dc.title.none.fl_str_mv PerfBound: Conserving Energy with Bounded Overheads in On/Off-Based HPC Interconnects
title PerfBound: Conserving Energy with Bounded Overheads in On/Off-Based HPC Interconnects
spellingShingle PerfBound: Conserving Energy with Bounded Overheads in On/Off-Based HPC Interconnects
Saravanan, Karthikeyan P.
High performance computing
Energy Efficient Interconnects
Energy Efficient Ethernet
Fast-Wake
Deep-Sleep
Supercomputadors
Àrees temàtiques de la UPC::Informàtica
title_short PerfBound: Conserving Energy with Bounded Overheads in On/Off-Based HPC Interconnects
title_full PerfBound: Conserving Energy with Bounded Overheads in On/Off-Based HPC Interconnects
title_fullStr PerfBound: Conserving Energy with Bounded Overheads in On/Off-Based HPC Interconnects
title_full_unstemmed PerfBound: Conserving Energy with Bounded Overheads in On/Off-Based HPC Interconnects
title_sort PerfBound: Conserving Energy with Bounded Overheads in On/Off-Based HPC Interconnects
dc.creator.none.fl_str_mv Saravanan, Karthikeyan P.
Carpenter, Paul Matthew
author Saravanan, Karthikeyan P.
author_facet Saravanan, Karthikeyan P.
Carpenter, Paul Matthew
author_role author
author2 Carpenter, Paul Matthew
author2_role author
dc.subject.none.fl_str_mv High performance computing
Energy Efficient Interconnects
Energy Efficient Ethernet
Fast-Wake
Deep-Sleep
Supercomputadors
Àrees temàtiques de la UPC::Informàtica
topic High performance computing
Energy Efficient Interconnects
Energy Efficient Ethernet
Fast-Wake
Deep-Sleep
Supercomputadors
Àrees temàtiques de la UPC::Informàtica
description Energy and power are key challenges in high-performance computing. System energy efficiency must be significantly improved, and this requires greater efficiency in all subcomponents. An important target of optimization is the interconnect, since network links are always on, consuming power even during idle periods. A large number of HPC machines have a primary interconnect based on Ethernet (about 40 percent of TOP500 machines), which, since 2010, has included support for saving power via Energy Efficient Ethernet (EEE). Nevertheless, it is unlikely that HPC interconnects would use these energy saving modes unless the performance overhead is known and small. This paper presents PerfBound, a self-contained technique to manage on/off-based networks such as EEE, minimizing interconnect link energy consumption subject to a bound on the performance degradation. PerfBound does not require changes to the applications and it uses only local information already available at switches and NICs without introducing additional communication messages, and is also compatible with multi-hop networks. PerfBound is evaluated using traces from a production supercomputer. For twelve out of fourteen applications, PerfBound has high energy savings, up to 70 percent for only 1 percent performance degradation. This paper also presents DynamicFastwake, which extends PerfBound to exploit multiple low-power states. DynamicFastwake achieves an energy-delay product 10 percent lower than the original PerfBound technique
publishDate 2018
dc.date.none.fl_str_mv 2018
2018-07-01
2018
2018-11-27
dc.type.none.fl_str_mv journal article
http://purl.org/coar/resource_type/c_6501
AM
http://purl.org/coar/version/c_ab4af688f83e57aa
dc.type.openaire.fl_str_mv info:eu-repo/semantics/article
format article
dc.identifier.none.fl_str_mv https://hdl.handle.net/2117/125138
https://dx.doi.org/10.1109/TC.2018.2790394
url https://hdl.handle.net/2117/125138
https://dx.doi.org/10.1109/TC.2018.2790394
dc.language.none.fl_str_mv Inglés
eng
language_invalid_str_mv Inglés
language eng
dc.relation.none.fl_str_mv European Commission http://dx.doi.org/10.13039/100011102 Seventh Framework Programme 287759 High Performance and Embedded Architecture and Compilation
dc.rights.none.fl_str_mv open access
http://purl.org/coar/access_right/c_abf2
dc.rights.openaire.fl_str_mv info:eu-repo/semantics/openAccess
rights_invalid_str_mv open access
http://purl.org/coar/access_right/c_abf2
eu_rights_str_mv openAccess
dc.format.none.fl_str_mv application/pdf
dc.publisher.none.fl_str_mv IEEE
publisher.none.fl_str_mv IEEE
dc.source.none.fl_str_mv reponame:UPCommons. Portal del coneixement obert de la UPC
instname:Universitat Politècnica de Catalunya (UPC)
instname_str Universitat Politècnica de Catalunya (UPC)
reponame_str UPCommons. Portal del coneixement obert de la UPC
collection UPCommons. Portal del coneixement obert de la UPC
repository.name.fl_str_mv
repository.mail.fl_str_mv
_version_ 1869413889076101120
score 15,300724

Similares en LA Referencia