ebook img

Parallel Computing Hits the Power Wall: Principles, Challenges, and a Survey of Solutions PDF

99 Pages·2019·3.208 MB·English
Save to my drive
Quick download
Download
Most books are stored in the elastic cloud where traffic is expensive. For this reason, we have a limit on daily download.

Preview Parallel Computing Hits the Power Wall: Principles, Challenges, and a Survey of Solutions

SPRINGER BRIEFS IN COMPUTER SCIENCE Arthur Francisco Lorenzon Antonio Carlos Schneider Beck Filho Parallel Computing Hits the Power Wall Principles, Challenges, and a Survey of Solutions 123 SpringerBriefs in Computer Science Serieseditors StanZdonik,BrownUniversity,Providence,RI,USA ShashiShekhar,UniversityofMinnesota,Minneapolis,MN,USA XindongWu,UniversityofVermont,Burlington,VT,USA LakhmiC.Jain,UniversityofSouthAustralia,Adelaide,SA,Australia DavidPadua,UniversityofIllinoisUrbana-Champaign,Urbana,IL,USA XueminShermanShen,UniversityofWaterloo,Waterloo,ON,Canada BorkoFurht,FloridaAtlanticUniversity,BocaRaton,FL,USA V.S.Subrahmanian,DepartmentofComputerScience,UniversityofMaryland, CollegePark,MD,USA MartialHebert,CarnegieMellonUniversity,Pittsburgh,PA,USA KatsushiIkeuchi,Meguro-ku,UniversityofTokyo,Tokyo,Japan BrunoSiciliano,DipartimentodiIngegneriaElettricaedelleTecnologie dell’Informazione,UniversitàdiNapoliFedericoII,Napoli,Italy SushilJajodia,GeorgeMasonUniversity,Fairfax,VA,USA NewtonLee,InstituteforEducationResearchandScholarships,LosAngeles, CA,USA SpringerBriefs present concise summaries of cutting-edge research and practical applicationsacrossawidespectrumoffields.Featuringcompactvolumesof50to 125pages,theseriescoversarangeofcontentfromprofessionaltoacademic. Typicaltopicsmightinclude: (cid:129) Atimelyreportofstate-of-theartanalyticaltechniques (cid:129) A bridge between new research results, as published in journal articles, and a contextualliteraturereview (cid:129) Asnapshotofahotoremergingtopic (cid:129) Anin-depthcasestudyorclinicalexample (cid:129) Apresentationofcoreconceptsthatstudentsmustunderstandinordertomake independentcontributions Briefsallowauthorstopresenttheirideasandreaderstoabsorbthemwithminimal time investment. Briefs will be published as part of Springer’s eBook collection, withmillionsofusersworldwide.Inaddition,Briefswillbeavailableforindividual print and electronic purchase. Briefs are characterized by fast, global electronic dissemination, standard publishing contracts, easy-to-use manuscript preparation and formatting guidelines, and expedited production schedules. We aim for pub- lication 8–12 weeks after acceptance. Both solicited and unsolicited manuscripts areconsideredforpublicationinthisseries. Moreinformationaboutthisseriesathttp://www.springer.com/series/10028 Arthur Francisco Lorenzon Antonio Carlos Schneider Beck Filho Parallel Computing Hits the Power Wall Principles, Challenges, and a Survey of Solutions 123 ArthurFranciscoLorenzon AntonioCarlosSchneiderBeckFilho DepartmentofComputerScience InstituteofInformatics,CampusdoVale FederalUniversityofPampa(UNIPAMPA) FederalUniversityofRioGrande Alegrete,RioGrandedoSul,Brazil doSul(UFRGS) PortoAlegre,RioGrandedoSul,Brazil ISSN2191-5768 ISSN2191-5776 (electronic) SpringerBriefsinComputerScience ISBN978-3-030-28718-4 ISBN978-3-030-28719-1 (eBook) https://doi.org/10.1007/978-3-030-28719-1 ©TheAuthor(s),underexclusivelicensetoSpringerNatureSwitzerlandAG2019 Thisworkissubjecttocopyright.AllrightsaresolelyandexclusivelylicensedbythePublisher,whether thewholeorpartofthematerialisconcerned,specificallytherightsoftranslation,reprinting,reuse ofillustrations,recitation,broadcasting,reproductiononmicrofilmsorinanyotherphysicalway,and transmissionorinformationstorageandretrieval,electronicadaptation,computersoftware,orbysimilar ordissimilarmethodologynowknownorhereafterdeveloped. Theuseofgeneraldescriptivenames,registerednames,trademarks,servicemarks,etc.inthispublication doesnotimply,evenintheabsenceofaspecificstatement,thatsuchnamesareexemptfromtherelevant protectivelawsandregulationsandthereforefreeforgeneraluse. Thepublisher,theauthors,andtheeditorsaresafetoassumethattheadviceandinformationinthisbook arebelievedtobetrueandaccurateatthedateofpublication.Neitherthepublishernortheauthorsor theeditorsgiveawarranty,expressorimplied,withrespecttothematerialcontainedhereinorforany errorsoromissionsthatmayhavebeenmade.Thepublisherremainsneutralwithregardtojurisdictional claimsinpublishedmapsandinstitutionalaffiliations. ThisSpringerimprintispublishedbytheregisteredcompanySpringerNatureSwitzerlandAG. Theregisteredcompanyaddressis:Gewerbestrasse11,6330Cham,Switzerland Thisbookisdedicatedtothememoryof MárciaCristinaandAuroraCera. Preface Efficiently exploiting thread-level parallelism from modern multicore systems has been challenging for software developers. While blindly increasing the number of threads may lead to performance gains, it can also result in a disproportionate increase in energy consumption. In the same way, optimization techniques for reducing energy consumption, such as DVFS and power gating, can lead to huge performance loss if used incorrectly. In this book, we present and discuss several techniques that address these challenges. We start by providing a brief theoreticalbackgroundonparallelcomputinginsoftwareandthesourcesofpower consumption. Then, we show how different parallel programming interfaces and communicationmodelsmayaffectenergyconsumptionindifferentways.Next,we discuss tuning techniques to adapt the number of threads/operating frequency to achievethebestcompromisebetweenperformanceandenergy.Wefinishthisbook withadetailedanalysisofarepresentativeexampleofanadaptiveapproach. Alegrete,Brazil ArthurFranciscoLorenzon PortoAlegre,Brazil AntonioCarlosSchneiderBeckFilho vii Acknowledgments TheauthorswouldliketothankthefriendsandcolleaguesatInformaticsInstitute oftheFederalUniversityofRioGrandedoSulandgiveaspecialthankstoallthe peopleintheEmbeddedSystemsLaboratory,whohavecontributedtothisresearch since2013. The authors would also like to thank the Brazilian research support agencies, FAPERGS,CAPES,andCNPq. ix Contents 1 RuntimeAdaptability:TheKeyforImprovingParallelApplications.. 1 1.1 Introduction .............................................................. 1 1.2 ScalabilityAnalysis...................................................... 3 1.2.1 VariablesInvolved............................................... 5 1.3 ThisBook ................................................................ 7 2 FundamentalConcepts....................................................... 9 2.1 ParallelComputinginSoftware......................................... 9 2.1.1 CommunicationModels......................................... 9 2.1.2 ParallelProgrammingInterfaces................................ 10 2.1.3 MulticoreArchitectures ......................................... 12 2.2 PowerandEnergyConsumption........................................ 13 2.2.1 DynamicVoltageandFrequencyScaling....................... 14 2.2.2 PowerGating..................................................... 15 3 TheImpactofParallelProgrammingInterfacesonEnergy............. 17 3.1 Methodology ............................................................. 17 3.1.1 Benchmarks...................................................... 17 3.1.2 MulticoreArchitectures ......................................... 19 3.1.3 ExecutionEnvironment.......................................... 20 3.1.4 Setup ............................................................. 22 3.2 Results.................................................................... 23 3.2.1 PerformanceandEnergyConsumption......................... 23 3.2.2 Energy-DelayProduct........................................... 30 3.2.3 TheInfluenceoftheStaticPowerConsumption............... 34 3.3 Discussion................................................................ 38 4 TuningParallelApplications................................................. 41 4.1 DesignSpaceExplorationofOptimizationTechniques................ 41 4.2 DynamicConcurrencyThrottling....................................... 42 4.2.1 ApproacheswithnoRuntimeAdaptation andnoTransparency............................................. 43 4.2.2 ApproacheswithRuntimeAdaptationand/orTransparency .. 45 xi xii Contents 4.3 DynamicVoltageandFrequencyScaling............................... 49 4.3.1 ApproacheswithnoRuntimeAdaptation andnoTransparency............................................. 49 4.3.2 ApproacheswithRuntimeAdaptationand/orTransparency .. 50 4.4 DCTandDVFS .......................................................... 51 4.4.1 ApproacheswithnoRuntimeAdaptation andnoTransparency............................................. 51 4.4.2 ApproacheswithRuntimeAdaptationand/orTransparency .. 52 5 CaseStudy:DCTwithAurora .............................................. 55 5.1 TheNeedforAdaptabilityandTransparency .......................... 55 5.2 Aurora:SeamlessOptimizationofOpenMPApplications............. 56 5.2.1 IntegrationtoOpenMP.......................................... 56 5.2.2 SearchAlgorithm................................................ 60 5.3 EvaluationofAurora..................................................... 63 5.3.1 Methodology..................................................... 63 5.3.2 Results............................................................ 66 6 Conclusions.................................................................... 79 References.......................................................................... 81

See more

The list of books you might like

Most books are stored in the elastic cloud where traffic is expensive. For this reason, we have a limit on daily download.