NDLI: Scalable Automatic Parallelization of Irregular Reductions on Shared Memory Multiprocessors

Please wait, while we are loading the content...

Scalable Automatic Parallelization of Irregular Reductions on Shared Memory Multiprocessors

Content Provider	Semantic Scholar
Author	Gutiérrez, Eladio Plata, Oscar G.
Copyright Year	2000
Abstract	ScalableAutomaticParallelizationofIrregularReductionsonSharedMemoryMultipro cessorsE.Guti�errezO.PlataE.L.ZapataDepartmentofComputerArchitecture,UniversityM�alagaP.O.Box4114,E-29080M�alaga,Spainfeladio,oscar,ezapatag@ac.um a.e sAbstractThispaperpresentsanewaral lelizationmethodforreductionsofarrayswithsubscriptedsubscriptsonscal-ablesharedmemorymultiprocessors.Themappingofcomputationsisbasedongroupingreductionloopit-erationsintosetsthatarefurtherdistributedacrossprocessors.Iterationsbelongingtothesamesetarechoseninsuchawaythatupdatedi erententriesthereductionarray.Thatis,loopdistributionim-pliesaconict-frewritedistributionoftherductionarray.Theiterationsetsaresetupbybuildingaloop-indexprefetchingarraythatal lowstoreorderoperlytheloopiterations.Theprosedmethoisgeneral,scalable,andeasytoimplementonacompiler.Inad-ditionitdealsinauniformwaywithoneandmultiplesubscriptarrays.Incaseofmultipleindirectionar-rays,writesontheeductionvectora ectingdi erentsetsaresolvedbyde ningconict-frsupersets.Aperformanceevaluationandcomparisonwithotherex-istingtechniquesispresented.Fromtheexperimentalresultsandperformanceanalysis,theproposedmethoappearsasaclearalternativetothearrayexpansionandprivatizedbu ertechniques,usualonstate-of-the-artparal lelizingcompilers,likePolarisorSUIF.Thescalabilityproblemthatthosetechniquesexhibitismiss-inginourmethod,asthememoryoverheadpresenteddoesnotdependonthenumberofprocessors.1Intro ductionParallelcomputationisincreasinglyb eingusedasameantoobtainenoughcomputingp owersolvelargescienti candengineeringproblems.Thehighcom-plexityoftheseproblems,aswellthecurrenthigh-p erformancearchitectures,stronglysupp ortthenecessityofdevelopingcompilertechniquestoe�-cientlymapsuchapplicationsontoaparallelarchitec-Thisworkassupp ortedbytheMinistryofEducationandScience(CICYT)ofSpain(TIC96-1125-C03)ture.However,manyoftheseco desexhibitirregularaccesspatternstothedata.Currentcommercialcom-pilers[17,18]areinsu�cientlydevelop edtodealwiththisdataaccesses,leadingtolowparallele�ciencieswhentheyareusedonsuchprograms.Reductionop erationsarefrequentlyfoundinthecoreoftheseapplications,asinthenextsimplelo op,doi=1,NA(f(i))=oprexpenddoThesymb oloprrepresentsanasso ciativeop eration(likesum,pro duct,maximum,minim...),andexprisanexpressionthatshouldnotcontainyreferencetothereductionarrayA().Thesubscriptf()de-p endsonthelo opindexi,andapp earsinrightandlefthandsideoftheassignmentsentence.Suchpatternistermedashistogramreduction[16,11].Duetothelo op-variantnatureofsubscriptarrayf(),lo op-carrieddep endencesmayb epresentatrun-time(ifitisnotap ermutationarray).Ageneralapproachtoparallelizeirregularco des,in-cludingreductions,isbasedontheinspector-executormodel.CHAOS[15]isawellknownimplementationofthismo delfordistributedmemorymachines.Irregu-larreductionsareparallelizedbyusinganinsp ectortolo catenon-lo caldataforeachpro cessor.Afterwards,anexecutormustgathernon-lo caldatab eforethere-duction,andmustscattertheresultsafterit.Thisstrategyintro ducesasigni cantoverheadduemainlytoitsgenerality,causedbtheinsp ector(communica-tionschedule,global-to-lo caladdresstranslation)andtheexecutor(communicationsandlo cal-to-globalad-dresstranslation).Thisoverheadisprop ortionaltothenumb erofnon-lo caldataaccesses.Inasharedmemorycontext,academicparallelizerslikePolaris[2]andSUIF[8]recognizeirregularreduc-tionsandparallelizethemusingthereplicatedbu erorthearrayexpansiontechniques.The rstmetho drepli-catesthereductionarrayonallpro cessors.Eachpro cessorcomputesap ortionofthereductiononitsprivatebu er.Lateraglobalreductionisobtainedby
File Format	PDF HTM / HTML
Alternate Webpage(s)	http://www.atc.ctima.uma.es/inv-des/inves/reports/2000/UMA-DAC-00-02.ps.gz
Language	English
Access Restriction	Open
Content Type	Text
Resource Type	Article

Central Library (ISO-9001:2015 Certified)
Indian Institute of Technology Kharagpur
Kharagpur, West Bengal, India | PIN - 721302

See location in the Map
03222 282435
Mail: support@ndl.gov.in

Sl.	Authority	Responsibilities	Communication Details
1	Ministry of Education (GoI), Department of Higher Education	Sanctioning Authority	https://www.education.gov.in/ict-initiatives
2	Indian Institute of Technology Kharagpur	Host Institute of the Project: The host institute of the project is responsible for providing infrastructure support and hosting the project	https://www.iitkgp.ac.in
3	National Digital Library of India Office, Indian Institute of Technology Kharagpur	The administrative and infrastructural headquarters of the project	Dr. B. Sutradhar bsutra@ndl.gov.in
4	Project PI / Joint PI	Principal Investigator and Joint Principal Investigators of the project	Dr. B. Sutradhar bsutra@ndl.gov.in Prof. Saswat Chakrabarti will be added soon
5	Website/Portal (Helpdesk)	Queries regarding NDLI and its services	support@ndl.gov.in
6	Contents and Copyright Issues	Queries related to content curation and copyright issues	content@ndl.gov.in
7	National Digital Library of India Club (NDLI Club)	Queries related to NDLI Club formation, support, user awareness program, seminar/symposium, collaboration, social media, promotion, and outreach	clubsupport@ndl.gov.in
8	Digital Preservation Centre (DPC)	Assistance with digitizing and archiving copyright-free printed books	dpc@ndl.gov.in
9	IDR Setup or Support	Queries related to establishment and support of Institutional Digital Repository (IDR) and IDR workshops	idr@ndl.gov.in