2002, Volume 4, Issue 5
Strategic Study of CAE >> 2002, Volume 4, Issue 5
Study on Data Locality and Its Optimization Technologies in Multiprocessor Systems
School of Computer Science , National University of Defense Technology , Changsha 410073, China
Next Previous
Abstract
As one of the most important issues in multiprocessor systems, data locality is a main direction of studies on such systems. Combining the recent research achievements by the authors with the current state of studies in this field, a systematic discussion on data locality and its optimization technologies in multiprocessor systems was brought forth. Driven by the deficiency of current locality measurement models, an enhanced model suitable for hierarchy parallel computer architectures was proposed. Besides, from the viewpoint of static locality optimization technologies, a data transformation framework based on projection and delaminating policy was designed. While on the side of dynamic locality optimization technologies, a dynamic page migration mechanism based on instantaneous access information was implemented. More discussion about both static and dynamic optimization policies was presented in this paper. In addition, in-depth investigation on memory consistency model ,which raised as a key problem in exploiting locality optimization, was also stressed and a new model centered on operating system——thread consistency model was put forward.
Keywords
computer ; multiprocessorsystem ; data locality ; locality measurement model ; data transformation framework ; page migration ; thread consistency model
Figures
图1
图2
图3
References
[ 1 ] WolfM , LamM .Adatalocalityoptimizingalgorithm[A].ProcSIGPLANConferenceonProgrammingLan guageDesignandImplementation[C], 1991.30～44
[ 2 ] MckinleyKS , CarrS , TsengCW .Improvingdatalo calitywithlooptransformation[J].ACMTransactionsonProgrammingLanguagesandSystems, July1996, 18 (4) :424～453
[ 3 ] KandemirM , ChoudharyA , RamanujamJ, etal.Amatrix basedapproachto globallocalityoptimization[J].JournalofParallelandDistributedComputing, 1999, 58:190～235
[ 4 ] SalisburgC , ChenZ , MelhemR .ModelingCommuni cationLocalityinMultiprocessors[J].JournalofParal lelandDistributedComputing5, 1999, 6:71～98
[ 5 ] TanakaA .Extensionoftheworkingsetformodelingspatiallocalityinprogrambehavior[A].Proc6thIn ternationalSymposiumonModeling, AnalysisandSimu lationofComputerandTelecommunicationSystems[C], 1998.27～35
[ 6 ] 夏 军, 杨学军, 曾丽芳, 等.基于投影分层技术的嵌套循环空间局部性优化方法[R].长沙:国防科技大学计算机学院, 2001
[ 7 ] EggersS , JeremiassenT .Eliminatingfalsesharing[A].ProcInternationalConferenceonParallelProcess ing[C], 1991, I :377～381
[ 8 ] TorrelasJ , LamM , HennesseyJ.Falsesharingandspatiallocalityinmulti processorcache[J].IEEETransactionsonComputers, 1994, 43 (6) :651～663
[ 9 ] ChowJ , SarkarV .Falsesharingeliminationbyselectionofruntimeschedulingparameters[A].Proc26thInter nationalConferenceonParallelProcessing[C], 1997
[10] CierniakM , LiW .Recoveringlogicaldataandcodestructures[R].TechniqueReport591, DepartmentofComputerScience, UniversityofRochester, 1995
[11] BlackD , GuptaA , WeberWD .Competitivemanage mentofdistributedsharedmemory[A].ProceedingsofCOMPCON [C], March1989.184～190
[12] VergheseB , DevienS , GuptaA , etal.Operatingsys temsupportforimprovingdatalocalityoncc numacom puteservers[A].ProceedingsoftheSeventhInterna tionalConferenceonArchitectualSupportforProgram mingLanguagesandOperatingSystems (ASPLOSVII) [C], October1996.279～289,
[13] BaylorS , EkanadhamK , JannJ, etal.Lazyhomemi grationfordistributedsharedmemorysystems[A ].ProceedingsofInternationalConferenceonHighPerfor manceComputing[C], December1997
[14] LaudonJ , LenoskiD .TheSGIorigin:AccNUMAhighlyscalableserver[A].Proceedingsofthe24thAn nualInternationalSymposiumonComputerArchitecture[C], May1997
[15] ChandraR , DevineS , VergheseB , etal.Schedulingandpagemigrationformultiprocessorcomputeservers[A ].ProceedingsArchitectureSupportforProgram mingLanguagesandOperatingSystems[C], October1994.12～24
[16] NikolopoulosD , PapatheodorouT , PolychronopoulosC , etal.Userleveldynamicpagemigrationformulti programmedshared memorymultiprocessors[A].Pro ceedingsofthe29thInternationalConferenceonParallelProcessing[C], August2000
[17] CarterJB , BennettJK , ZwaenepoelW .Techniquesforreducingconsistency relatedinformationindistribut edsharedmemorysystems[J].ACMTransactinosonComputerSystems, August1995, 13 (3) :205～243
[18] YangXuejun, DaiHuadong.Operationsystem centricmemoryconsistencymodel———threadconsistencymodel[A], TheFourthInternationalWorkshoponAdvancedParallelProcessingTechnologies (APPT’01) [C ], September2001.26～36
京公网安备 11010502051620号
