Thermal Safe Power (TSP): Efficient Power Budgeting for Heterogeneous Manycore Systems in Dark Silicon
Chip manufacturers provide the Thermal Design Power (TDP) for a specific chip. The cooling solution is designed to dissipate this power level. But because TDP is not necessarily the maximum power that can be applied, chips are operated with Dynamic Thermal Management (DTM) techniques. To avoid exces...
Gespeichert in:
Veröffentlicht in: | IEEE transactions on computers 2017-01, Vol.66 (1), p.147-162 |
---|---|
Hauptverfasser: | , , , , , |
Format: | Artikel |
Sprache: | eng |
Schlagworte: | |
Online-Zugang: | Volltext bestellen |
Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
container_end_page | 162 |
---|---|
container_issue | 1 |
container_start_page | 147 |
container_title | IEEE transactions on computers |
container_volume | 66 |
creator | Pagani, Santiago Khdr, Heba Jian-Jia Chen Shafique, Muhammad Minming Li Henkel, Jorg |
description | Chip manufacturers provide the Thermal Design Power (TDP) for a specific chip. The cooling solution is designed to dissipate this power level. But because TDP is not necessarily the maximum power that can be applied, chips are operated with Dynamic Thermal Management (DTM) techniques. To avoid excessive triggers of DTM, usually, system designers also use TDP as power constraint. However, using a single and constant value as power constraint, e.g., TDP, can result in significant performance losses in homogeneous and heterogeneous manycore systems. Having better power budgeting techniques is a major step towards dealing with the dark silicon problem. This paper presents a new power budget concept, called Thermal Safe Power (TSP), which is an abstraction that provides safe power and power density constraints as a function of the number of simultaneously active cores. Executing cores at any power consumption below TSP ensures that DTM is not triggered. TSP can be computed offline for the worst cases, or online for a particular mapping of cores. TSP can also serve as a fundamental tool for guiding task partitioning and core mapping decisions, specially when core heterogeneity or timing guarantees are involved. Moreover, TSP results in dark silicon estimations which are less pessimistic than estimations using constant power budgets. |
doi_str_mv | 10.1109/TC.2016.2564969 |
format | Article |
fullrecord | <record><control><sourceid>proquest_RIE</sourceid><recordid>TN_cdi_proquest_journals_1848276973</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><ieee_id>7466857</ieee_id><sourcerecordid>1848276973</sourcerecordid><originalsourceid>FETCH-LOGICAL-c289t-6a487c1248e8adc689d23ea39b294666d4d162c6b7da1a2f1269a379d4d1d01b3</originalsourceid><addsrcrecordid>eNo9kMFLwzAUh4MoOKdnD14CXvTQLUnbNPGmdTph4qD1HLL0dWZuzUxaZP-9HRueHvz4fu89PoSuKRlRSuS4zEeMUD5iKU8klydoQNM0i6RM-SkaEEJFJOOEnKOLEFaEEM6IHKC6_AK_0Wtc6Brw3P2Cx3dlMb9_wJO6tsZC0x7jp65aQmubJa6dx1NowbslNOC6gN91szPOAy52oYVNwLbBz9p_48KurXHNJTqr9TrA1XEO0efLpMyn0ezj9S1_nEWGCdlGXCciM5QlAoSuDBeyYjHoWC6YTDjnVVJRzgxfZJWmmtWUcanjTO7zitBFPES3h71b7346CK1auc43_UlFRSJYxmUW99T4QBnvQvBQq623G-13ihK1l6nKXO1lqqPMvnFzaFgA-Kez_ieRZvEfAUdvLQ</addsrcrecordid><sourcetype>Aggregation Database</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>1848276973</pqid></control><display><type>article</type><title>Thermal Safe Power (TSP): Efficient Power Budgeting for Heterogeneous Manycore Systems in Dark Silicon</title><source>IEEE Electronic Library (IEL)</source><creator>Pagani, Santiago ; Khdr, Heba ; Jian-Jia Chen ; Shafique, Muhammad ; Minming Li ; Henkel, Jorg</creator><creatorcontrib>Pagani, Santiago ; Khdr, Heba ; Jian-Jia Chen ; Shafique, Muhammad ; Minming Li ; Henkel, Jorg</creatorcontrib><description>Chip manufacturers provide the Thermal Design Power (TDP) for a specific chip. The cooling solution is designed to dissipate this power level. But because TDP is not necessarily the maximum power that can be applied, chips are operated with Dynamic Thermal Management (DTM) techniques. To avoid excessive triggers of DTM, usually, system designers also use TDP as power constraint. However, using a single and constant value as power constraint, e.g., TDP, can result in significant performance losses in homogeneous and heterogeneous manycore systems. Having better power budgeting techniques is a major step towards dealing with the dark silicon problem. This paper presents a new power budget concept, called Thermal Safe Power (TSP), which is an abstraction that provides safe power and power density constraints as a function of the number of simultaneously active cores. Executing cores at any power consumption below TSP ensures that DTM is not triggered. TSP can be computed offline for the worst cases, or online for a particular mapping of cores. TSP can also serve as a fundamental tool for guiding task partitioning and core mapping decisions, specially when core heterogeneity or timing guarantees are involved. Moreover, TSP results in dark silicon estimations which are less pessimistic than estimations using constant power budgets.</description><identifier>ISSN: 0018-9340</identifier><identifier>EISSN: 1557-9956</identifier><identifier>DOI: 10.1109/TC.2016.2564969</identifier><identifier>CODEN: ITCOB4</identifier><language>eng</language><publisher>New York: IEEE</publisher><subject>Budgeting ; Chips (electronics) ; dark silicon ; heterogeneity ; Multicore processing ; Power demand ; Power management ; Silicon ; Steady-state ; Temperature dependence ; Temperature measurement ; Thermal design ; thermal design power (TDP) ; Thermal management ; Thermal safe power (TSP)</subject><ispartof>IEEE transactions on computers, 2017-01, Vol.66 (1), p.147-162</ispartof><rights>Copyright The Institute of Electrical and Electronics Engineers, Inc. (IEEE) 2017</rights><lds50>peer_reviewed</lds50><woscitedreferencessubscribed>false</woscitedreferencessubscribed><citedby>FETCH-LOGICAL-c289t-6a487c1248e8adc689d23ea39b294666d4d162c6b7da1a2f1269a379d4d1d01b3</citedby><cites>FETCH-LOGICAL-c289t-6a487c1248e8adc689d23ea39b294666d4d162c6b7da1a2f1269a379d4d1d01b3</cites><orcidid>0000-0002-5088-0347</orcidid></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><linktohtml>$$Uhttps://ieeexplore.ieee.org/document/7466857$$EHTML$$P50$$Gieee$$H</linktohtml><link.rule.ids>315,781,785,797,27929,27930,54763</link.rule.ids><linktorsrc>$$Uhttps://ieeexplore.ieee.org/document/7466857$$EView_record_in_IEEE$$FView_record_in_$$GIEEE</linktorsrc></links><search><creatorcontrib>Pagani, Santiago</creatorcontrib><creatorcontrib>Khdr, Heba</creatorcontrib><creatorcontrib>Jian-Jia Chen</creatorcontrib><creatorcontrib>Shafique, Muhammad</creatorcontrib><creatorcontrib>Minming Li</creatorcontrib><creatorcontrib>Henkel, Jorg</creatorcontrib><title>Thermal Safe Power (TSP): Efficient Power Budgeting for Heterogeneous Manycore Systems in Dark Silicon</title><title>IEEE transactions on computers</title><addtitle>TC</addtitle><description>Chip manufacturers provide the Thermal Design Power (TDP) for a specific chip. The cooling solution is designed to dissipate this power level. But because TDP is not necessarily the maximum power that can be applied, chips are operated with Dynamic Thermal Management (DTM) techniques. To avoid excessive triggers of DTM, usually, system designers also use TDP as power constraint. However, using a single and constant value as power constraint, e.g., TDP, can result in significant performance losses in homogeneous and heterogeneous manycore systems. Having better power budgeting techniques is a major step towards dealing with the dark silicon problem. This paper presents a new power budget concept, called Thermal Safe Power (TSP), which is an abstraction that provides safe power and power density constraints as a function of the number of simultaneously active cores. Executing cores at any power consumption below TSP ensures that DTM is not triggered. TSP can be computed offline for the worst cases, or online for a particular mapping of cores. TSP can also serve as a fundamental tool for guiding task partitioning and core mapping decisions, specially when core heterogeneity or timing guarantees are involved. Moreover, TSP results in dark silicon estimations which are less pessimistic than estimations using constant power budgets.</description><subject>Budgeting</subject><subject>Chips (electronics)</subject><subject>dark silicon</subject><subject>heterogeneity</subject><subject>Multicore processing</subject><subject>Power demand</subject><subject>Power management</subject><subject>Silicon</subject><subject>Steady-state</subject><subject>Temperature dependence</subject><subject>Temperature measurement</subject><subject>Thermal design</subject><subject>thermal design power (TDP)</subject><subject>Thermal management</subject><subject>Thermal safe power (TSP)</subject><issn>0018-9340</issn><issn>1557-9956</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2017</creationdate><recordtype>article</recordtype><sourceid>RIE</sourceid><recordid>eNo9kMFLwzAUh4MoOKdnD14CXvTQLUnbNPGmdTph4qD1HLL0dWZuzUxaZP-9HRueHvz4fu89PoSuKRlRSuS4zEeMUD5iKU8klydoQNM0i6RM-SkaEEJFJOOEnKOLEFaEEM6IHKC6_AK_0Wtc6Brw3P2Cx3dlMb9_wJO6tsZC0x7jp65aQmubJa6dx1NowbslNOC6gN91szPOAy52oYVNwLbBz9p_48KurXHNJTqr9TrA1XEO0efLpMyn0ezj9S1_nEWGCdlGXCciM5QlAoSuDBeyYjHoWC6YTDjnVVJRzgxfZJWmmtWUcanjTO7zitBFPES3h71b7346CK1auc43_UlFRSJYxmUW99T4QBnvQvBQq623G-13ihK1l6nKXO1lqqPMvnFzaFgA-Kez_ieRZvEfAUdvLQ</recordid><startdate>20170101</startdate><enddate>20170101</enddate><creator>Pagani, Santiago</creator><creator>Khdr, Heba</creator><creator>Jian-Jia Chen</creator><creator>Shafique, Muhammad</creator><creator>Minming Li</creator><creator>Henkel, Jorg</creator><general>IEEE</general><general>The Institute of Electrical and Electronics Engineers, Inc. (IEEE)</general><scope>97E</scope><scope>RIA</scope><scope>RIE</scope><scope>AAYXX</scope><scope>CITATION</scope><scope>7SC</scope><scope>7SP</scope><scope>8FD</scope><scope>JQ2</scope><scope>L7M</scope><scope>L~C</scope><scope>L~D</scope><orcidid>https://orcid.org/0000-0002-5088-0347</orcidid></search><sort><creationdate>20170101</creationdate><title>Thermal Safe Power (TSP): Efficient Power Budgeting for Heterogeneous Manycore Systems in Dark Silicon</title><author>Pagani, Santiago ; Khdr, Heba ; Jian-Jia Chen ; Shafique, Muhammad ; Minming Li ; Henkel, Jorg</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-c289t-6a487c1248e8adc689d23ea39b294666d4d162c6b7da1a2f1269a379d4d1d01b3</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2017</creationdate><topic>Budgeting</topic><topic>Chips (electronics)</topic><topic>dark silicon</topic><topic>heterogeneity</topic><topic>Multicore processing</topic><topic>Power demand</topic><topic>Power management</topic><topic>Silicon</topic><topic>Steady-state</topic><topic>Temperature dependence</topic><topic>Temperature measurement</topic><topic>Thermal design</topic><topic>thermal design power (TDP)</topic><topic>Thermal management</topic><topic>Thermal safe power (TSP)</topic><toplevel>peer_reviewed</toplevel><toplevel>online_resources</toplevel><creatorcontrib>Pagani, Santiago</creatorcontrib><creatorcontrib>Khdr, Heba</creatorcontrib><creatorcontrib>Jian-Jia Chen</creatorcontrib><creatorcontrib>Shafique, Muhammad</creatorcontrib><creatorcontrib>Minming Li</creatorcontrib><creatorcontrib>Henkel, Jorg</creatorcontrib><collection>IEEE All-Society Periodicals Package (ASPP) 2005–Present</collection><collection>IEEE All-Society Periodicals Package (ASPP) 1998-Present</collection><collection>IEEE Electronic Library (IEL)</collection><collection>CrossRef</collection><collection>Computer and Information Systems Abstracts</collection><collection>Electronics & Communications Abstracts</collection><collection>Technology Research Database</collection><collection>ProQuest Computer Science Collection</collection><collection>Advanced Technologies Database with Aerospace</collection><collection>Computer and Information Systems Abstracts Academic</collection><collection>Computer and Information Systems Abstracts Professional</collection><jtitle>IEEE transactions on computers</jtitle></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext_linktorsrc</fulltext></delivery><addata><au>Pagani, Santiago</au><au>Khdr, Heba</au><au>Jian-Jia Chen</au><au>Shafique, Muhammad</au><au>Minming Li</au><au>Henkel, Jorg</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Thermal Safe Power (TSP): Efficient Power Budgeting for Heterogeneous Manycore Systems in Dark Silicon</atitle><jtitle>IEEE transactions on computers</jtitle><stitle>TC</stitle><date>2017-01-01</date><risdate>2017</risdate><volume>66</volume><issue>1</issue><spage>147</spage><epage>162</epage><pages>147-162</pages><issn>0018-9340</issn><eissn>1557-9956</eissn><coden>ITCOB4</coden><abstract>Chip manufacturers provide the Thermal Design Power (TDP) for a specific chip. The cooling solution is designed to dissipate this power level. But because TDP is not necessarily the maximum power that can be applied, chips are operated with Dynamic Thermal Management (DTM) techniques. To avoid excessive triggers of DTM, usually, system designers also use TDP as power constraint. However, using a single and constant value as power constraint, e.g., TDP, can result in significant performance losses in homogeneous and heterogeneous manycore systems. Having better power budgeting techniques is a major step towards dealing with the dark silicon problem. This paper presents a new power budget concept, called Thermal Safe Power (TSP), which is an abstraction that provides safe power and power density constraints as a function of the number of simultaneously active cores. Executing cores at any power consumption below TSP ensures that DTM is not triggered. TSP can be computed offline for the worst cases, or online for a particular mapping of cores. TSP can also serve as a fundamental tool for guiding task partitioning and core mapping decisions, specially when core heterogeneity or timing guarantees are involved. Moreover, TSP results in dark silicon estimations which are less pessimistic than estimations using constant power budgets.</abstract><cop>New York</cop><pub>IEEE</pub><doi>10.1109/TC.2016.2564969</doi><tpages>16</tpages><orcidid>https://orcid.org/0000-0002-5088-0347</orcidid></addata></record> |
fulltext | fulltext_linktorsrc |
identifier | ISSN: 0018-9340 |
ispartof | IEEE transactions on computers, 2017-01, Vol.66 (1), p.147-162 |
issn | 0018-9340 1557-9956 |
language | eng |
recordid | cdi_proquest_journals_1848276973 |
source | IEEE Electronic Library (IEL) |
subjects | Budgeting Chips (electronics) dark silicon heterogeneity Multicore processing Power demand Power management Silicon Steady-state Temperature dependence Temperature measurement Thermal design thermal design power (TDP) Thermal management Thermal safe power (TSP) |
title | Thermal Safe Power (TSP): Efficient Power Budgeting for Heterogeneous Manycore Systems in Dark Silicon |
url | https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2024-12-13T10%3A15%3A51IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-proquest_RIE&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=Thermal%20Safe%20Power%20(TSP):%20Efficient%20Power%20Budgeting%20for%20Heterogeneous%20Manycore%20Systems%20in%20Dark%20Silicon&rft.jtitle=IEEE%20transactions%20on%20computers&rft.au=Pagani,%20Santiago&rft.date=2017-01-01&rft.volume=66&rft.issue=1&rft.spage=147&rft.epage=162&rft.pages=147-162&rft.issn=0018-9340&rft.eissn=1557-9956&rft.coden=ITCOB4&rft_id=info:doi/10.1109/TC.2016.2564969&rft_dat=%3Cproquest_RIE%3E1848276973%3C/proquest_RIE%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_pqid=1848276973&rft_id=info:pmid/&rft_ieee_id=7466857&rfr_iscdi=true |