thek-aryn-cubestructure. 1
|
|
|
- Corey Robbins
- 10 years ago
- Views:
Transcription
1 DDE:AModiedDimensionExchangeMethod forloadbalancingink-aryn-cubes StateUniversityofNewYorkatBualo DepartmentofComputerScience Min-YouWuandWeiShu algorithmforthehypercubestructure.ithasbeengeneralizedtok-aryn-cubes.however,the k-aryn-cubealgorithmmusttakemanyiterationstoconvergetoabalancedstate.inthispaper, Abstract Thedimensionexchangemethod(DEM)wasinitiallyproposedasaload-balancing Bualo,NY14260 balancestheloaddirectlywithoutiterativelyexchangingtheload.itisabletobalancetheload moreaccuratelyandmuchfaster.1.introduction (DDE)method,takesloadaverageineverydimensiontoeliminateunnecessaryloadexchange.It weproposeadirectmethodtomodifydem.thenewalgorithm,directdimensionexchange DEMissuperiortootherschedulingmethods[7].DEMforthehypercubenetworkisasimple memorymachines.theexperimentcarriedbywillebeek-lemairandreevesconformedthat algorithm.loadbalancingisperformediterativelyineachofthelogndimensions,inwhichonly rithmforthehypercubestructure[5,1].itbalancestheloadforindependenttasksondistributed nodepairsexchangetheirloadinformationandattempttoaveragethenumberoftasks.aftera sweep(logniterations),theloadisbalanced. Thedimensionexchangemethod(DEM)wasinitiallyproposedasafullyload-balancingalgo- ittakesmanysweepstoconvergetothebalancedload.hosseinietal.extendeditforarbitrary dimensionexchange(gde)method[9].thegdemethodwasextendedtothek-aryn-cube network[10].becauseanodeexchangesworkloadwithonlyoneofitsneighboratatime,gde structuresusingthetechniqueofedge-coloringofgraphs[3].xuandlauproposedthegeneralized isnotabletoreachthebalancedstateinonesweep.thenumberofsweepsforconvergenceis linearlyproportionaltothenumberofnodesinachain,andhencetothedimensionorderkof Unfortunately,whenDEMappliestootherstructures,suchasthemeshorthek-aryn-cube, thek-aryn-cubestructure. 1
2 beappliedtotwoormoredimensionstobalancetheloadforthemesh,thetorus,andthek-ary n-cube. orunderloadedandsubsequentlyexchangeworkloadwithothernodes.theddemethodcan loadinachainisfullybalancedbyutilizinginformationofthetotalnumberoftasks,whichcan method.unlikeiterativealgorithms,thisdirectmethodcanbalancetheloadinonesweep.the beeasilyobtainedbyasumreduction.eachnodeinthechainknowswhetheritisoverloaded Wepresentadirectmethodforthek-aryn-cube,calledtheDirectDimensionExchange(DDE) respectively.thealgorithmforthek-aryn-cubeispresentedinsection5.insection6,thedirect Then,thedirectmethodforthechainandtheringstructuresisdescribedinsections3and4, methodiscomparedtothegdemethod.section7concludesthepaper. Thispaperisorganizedasfollows.Section2brieyreviewstheDEMandtheGDEalgorithms. Toachievethisgoal,anestimationofthetaskexecutiontimeisneeded,whichcanbedoneeither byaprogrammerorbyacompiler.sometimestheestimationcanbeapplication-specic,and sometimesitisimpossibletoobtainsuchanestimation.duetothesediculties,eachtaskis Thegoalofloadbalancingistoscheduleworkssothateachprocessorhasthesameworkload. 2.TheDEMandGDEAlgorithms presumedtorequiretheequalexecutiontimeandthegoalofthealgorithmistoscheduletasks sothateachprocessorhasthesamenumberoftasks. algorithmistoredistributetaskssothatthenumberoftasksineachnodeisequal.assumethe sumofwiofallnodescanbeevenlydividedbyn.theaveragenumberoftaskswavgiscalculated computingnodesareconnectedbyagiventopology.eachnodeihaswitasks.ascheduling Theschedulingproblemcanbedescribedasfollows.Inaparallelordistributedsystem,N andthencombinedtoformlargerdomainsuntilultimatelytheentiresystemisbalanced.the Eachnodeshouldhavewavgtasksafterscheduling. \integerversion"ofdemisdescribedinfigure1.allnodepairsintherstdimensionwhose DEMwasdesignedforthehypercubestructure.InDEM,smalldomainsarebalancedrst wavg=pn?1 i=0wi addressesdierinonlytheleastsignicantbitbalancetheloadbetweenthemselves.next,all N: nodehasbalanceditsloadwitheachofitsneighbors. nodepairsintheseconddimensionbalancetheloadbetweenthemselves,andsoforth,untileach AfterexecutionoftheDEMalgorithm,theloaddierence D=max(wi)?min(wi) 2
3 DEM forl=0ton?1 wi=(d(wi+wj)=2eifwi>wj if(wj?wi)>1,receiveb(wj?wi)=2ctasksfromnodej if(wi?wj)>1,sendb(wi?wj)=2ctaskstonodej nodeiexchangeswithnodejthecurrentvaluesofwiandwj,wherej=i2l GDE b(wi+wj)=2cotherwise while(notterminate) forl=1toc foredgecoloredlconnectingnodesiandj Figure1:TheDEMalgorithmforthehypercube. nodeiexchangeswithnodejthecurrentvaluesofwiandwj if(wi?wj)>1,sendb(wi?wj)ctaskstonodej wi=(d(1?)wi+wjeifwi>wj if(wj?wi)>1,receiveb(wj?wi)ctasksfromnodej isboundedbyn,thedimensionofthehypercube[3].thenumberofcommunicationstepsofthe DEMalgorithmis3n[7]. Figure2:TheGDEalgorithmforthek-aryn-cube. b(1?)wi+wjcotherwise graph.the\integerversion"ofthealgorithmisshowninfigure2.anodenishesacomplete Forthehypercube,theoptimal=12,andGDEisequivalenttotheoriginalDEMalgorithm.For c=2nifkisanevennumber.theterminationconditionisthatthedierenceofthenumberof sweepaftercconsecutiveexchangeoperations,wherecisthenumberofcolors.ink-aryn-cubes, theexchangeparameter.thevaluevariesfordierenttopologiesanddierentnetworksizes. tasksbetweenneighboringnodesislessthanorequaltoone.theconvergenceratedependson TheGDEalgorithmoperatesoncolorgraphsderivedfromedge-coloringofthegivensystem loaddierencebetweenanypairofnodesisboundedbynk=2.theconvergenceratedecreases othertopologies,istobeoptimizedtomaximizetheconvergencerate.forthek-aryn-cube,the whenthedimensionorderkincreases.thereisnocommunicationconictinthisalgorithm. topologyby\folding"themeshineachdimensiondlogmetimes[7].thismethodcouldbeapplied Willebeek-LeMairandReevessuggestedanotherapproachtoextendDEMtoanMMmesh 3
4 pairswouldnolongerbedirectlylinkedtooneanotherandcommunicationswouldconict. tok-aryn-cubestoo.theloaddierenceisboundedbyndlogke.however,inthisapproach,node method.theworkloadinachaincanbebalanceddirectly.thebasicideaistocalculatethe totalnumberoftasksinthechainandtheaveragenumberoftaskspernode.thus,nodesinthe chaincanexchangetaskstobalancetheload. InsteadofusingtheGDEmethodwhichbalancestheloaditeratively,weproposeadirect 3.TheDDEMethodfortheChain DDE-chain Letwibethenumberoftasksinnodei,wherei=0;1;:::;k?1. 2.AverageLoadCalculation:T=W0,wavg=bT=kc,andR=Tmodk,whereTisthe 1.GlobalInformationCollection:Performthescanwithsumoperationofwi: 3.QuotaCalculation:Thequotaofeachnodeqiiscomputed: totalnumberoftasks. Wi=k?1 Xl=iwl qi=(wavg+1ifi<r 4.FlowCalculation:xi?1;i=Qi?Wi,fori=1;2;:::;k?1,wherexi;jistheowon Also,anaccumulationquotaforeachnodeiscomputed: wavgotherwise edge(i;j). Qi=k?1 Figure3:TheDDEalgorithmforthechain. Xl=iql thenodeweightwi(i=0;1;:::;k?1)andoutputsthecalculatedowxi?1;i(i=1;2;:::;k?1) foreveryedgeinthechain.therststepistoobtainthetotalnumberoftasksinthechainby usingthescanwithsumoperationfromnodek?1tonode0,wherekisthelengthofthechain. EachnoderecordsapartialsumWi=Pk?1 oftaskspernodeatnode0.ifthenumberoftaskscannotbeevenlydividedbyk,theremaining TheDDEalgorithmforthechainshowninFigure3isits\integerversion."Ittakesasinput l=iwl.thesecondstepcalculatestheaveragenumber 4
5 RtasksaredistributedtotherstRnodessothattheyhaveonemoretaskthantheothers. ThevaluesofwavgandRarebroadcasttoeverynode.Inthethirdstep,eachnodecalculatesits asitsquota. EachnodekeepsrecordsofQi,Wi,Qj,andWj,wherej=i+1.Inthefourthstep,theow iscalculatedbytakingdierencebetweenqiandwi.nodeicalculatesxi?1;iandxi;i+1.when quota.theaccumulationquotaqicanbecalculateddirectlyasfollows: theowisavailable,theworkloadisexchangedsothateachnodehasthesamenumberoftasks Qi=wavg(k?i)+min(0;R?i): Example1: readytobescheduled.valuesofwiarecalculatedinstep1.node0calculatesthevalueofwavg andr: shownbelow: Then,eachnodecalculatesthevalueofQiinstep3.Thevaluesofwi,Wi,Qi,andxi?1;iareas AnexampleisshowninFigure4.Atthebeginningofscheduling,eachnodehaswitasks iwiwiqixi?1;i wavg=4;r=5: { i=0 94i=1 76i= ?1 5i=3 11i=4 42i=5 62i=6 11 Aftertaskexchange,nodes0{4havevetaskseach,andnodes5{7havefourtaskseach. Figure4:ExampleforDDE-chain. i=7 5 toitsquota. Lemma1:AfterexecutionofDDEandtaskexchange,thenumberoftasksineachnodeisequal 5
6 Becausexi?1;i=Qi?Wi;xi;i+1=Qi+1?Wi+1;Wi+1=Wi?wi;andQi+1=Qi?qi Proof:AfterexecutionofDDEandtaskexchange,thenumberoftasksinnodeiis w0i=wi+(qi?wi)?(qi+1?wi+1)=qi?qi+1=qi w0i=wi+xi?1;i?xi;i+1 stepsinstep4isatmostk.therefore,thetotalnumberofcommunicationstepsofthisalgorithm andapplyingthetwaalgorithmin[6].thus,thetotalnumberofcommunicationstepsofthis isnomorethan3k.thisalgorithmcanbefurtherimprovedbyselectingnodek/2astheroot algorithmcanbereducedto2k.whentisevenlydividedbyk,thisalgorithmminimizesthe Inthisalgorithm,steps1and2spend2kcommunicationsteps.Thenumberofcommunication 2 Receive-before-send totalnumberoftasktransfersandthetotalnumberofcommunications.thisalgorithmalso maximizeslocality.thatis,itminimizesthenumberoftasksthataremigratedtoothernodes. Fornodei exchangealgorithms.therstone,calledreceive-before-send,isshowninfigure5. 1.ifi>0andxi?1;i>0,waittoreceivexi?1;itasksfromnodei?1 TheworkloadisexchangedaccordingtotheowgeneratedbyDDE.Therearetwotask- 2.ifi<k?1andxi;i+1<0,waittoreceivejxi;i+1jtasksfromnodei+1 3.ifi>0andxi?1;i<0,sendjxi?1;ijtaskstonodei?1 nicationstepstonish: 4.ifi<k?1andxi;i+1>0,sendxi;i+1taskstonodei+1 Usingthereceive-before-sendalgorithm,theloadexchangeinExample1takesfourcommu- (2) (1) node0tonode1,node5tonode6,node7tonode6 node1tonode2 Figure5:Taskexchange:receive-before-send. (3) (4) node2tonode3 node3tonode4 6
7 while(ai6=0orbi6=0) letai=xi?1;i;bi=xi;i+1 Send-before-receive Fornodei 4.ifi<k?1andbi<0andreceivedjbijtasksfromnodei+1,andletwi=wi?bi,bi=0 3.ifi>0andai>0andreceivedaitasksfromnodei?1,andletwi=wi+ai,ai=0 2.ifi<k?1and(wi>bi>0)sendbitaskstonodei+1,andletwi=wi?bi,bi=0 1.ifi>0and(wi>?ai>0)sendjaijtaskstonodei?1,andletwi=wi+ai,ai=0 infigure6.inthisalgorithm,anodecanstartsendingmessagesoutbeforeithasreceivedan incomingmessage.thecommunicationtimeandprocessoridletimecanbereduced.ittakes beforesendingoutmessages.byrelaxingthisconstraint,asend-before-receivealgorithmisshown onlytwocommunicationstepsforexample1: Inthereceive-before-sendalgorithm,eachnodemustreceiveanincomingmessage,ifany, Figure6:Taskexchange:send-before-receive. 1) 2) node0tonode1,node1tonode2,node3tonode4, node5tonode6,node7tonode6 before-sendorsend-before-receivealgorithmsisatrade-obetweencommunicationtimeand othernodesandthenreceivetasksfromothers.therefore,thedecisiononuseofthereceive- taskstoothernodes.butinthesend-before-receivealgorithm,anodemaysendlocaltasksto before-sendalgorithm,anodecankeepthemaximumnumberoflocaltasksandsendnon-local Thesend-before-receivealgorithmmayhavesomenegativeimpactinlocality.Inthereceive- node2tonode3 locality. oncommunicationtimecanoftenbeignored.therecursivedoublingalgorithm[2]cantake advantageofthepipelineeectofwormholeroutingwhileavoidingchannelcontention.this algorithmorganizesthenodesinachaintoatree.anexampleofeightnodesisshownin communicationsteps. Figure7.ApplyingtheTWAalgorithmin[6]tothetree,theloadcanbebalancedwithin4logk Mostmassivelyparallelcomputersusewormholeroutingwithwhichtheeectofpathlength 7
8 i=0 i=6 i=4 i=5 i=2 i=3 i=1 4.TheDDEAlgorithmfortheRing Figure7:Thetreeforrecursivedoubling. i=7 however,thecommunicationmaynotbeminimal.byutilizingtheend-roundedge,communicationcouldbereduced.wedescribeanalgorithmtominimizethetotalnumberoftaskstransferred. Thealgorithmisderivedfromtheminimumcostowalgorithm[4]andshowninFigure8.Inthis gorithmcanbeappliedtotheringbyignoringtheend-roundedge.theloadcanbebalanced, algorithm,aninitialsolutionisobtainedbyusingdde-chainwithoutconsideringtheend-round Aringcanbeobtainedbyaddinganend-roundconnectiontoachain.TheDDE-chainal- x0;1;x1;2;:::;xk?2;k?1,wherexi;jistheowonedge(i;j).letxk?1;0be0. DDE-ring ApplyDDE-chaintotheringwithoutconsideringtheend-roundedge(k?1;0)toobtain Iftheowisclockwise,xi;jispositive;otherwise,itisnegative. Letnpbethenumberofedgeswithxi;j>0,nnthenumberofedgeswithxi;j<0,andnzthe numberofedgeswithxi;j=0. 1.Ifnn+nz?np<0,letxmbethemthlargestxi;jfromallxi;j>0;andifnp+nz?nn<0, 2.Foreachedge,xi;j=xi;j?xm. letxmbethemthsmallestxi;jfromallxi;j<0,wherem=dk=2e. Figure8:TheDDEalgorithmforthering. 8
9 edge.then,anaugmentationisappliedtoobtainanoptimalsolution.thecomplexityofthis algorithmiso(klogk). ofdde-ring.here,weletx?1;0=xk?1;0. Wecanuseeitherthereceive-before-sendorsend-before-receivealgorithmfortaskexchange negativecost.therefore,thenetworkowisofminimumcost[4]. numberoftaskstransferred. Lemma2:AfterexecutionofDDE-ring,thetotalnetworkowisofminimumcost. Proof:Ifnp+nz?nn0andnn+nz?np0,thereisnoowaugmentingcyclewith Thefollowinglemmashowsthatthisalgorithmminimizesthetotalcostofow,thatis,the Notethatn0n+n0z+n0p=k.Becauseofm=dk=2e, Then, Ifnn+nz?np<0,aftermodicationofxi;j=xi;j?xm,wehave n0z+n0p?n0nm?n0n=m?(k?n0z?n0p)2m?k n0z+n0p?n0n2m?k0 n0z+n0pm Then,n0n+n0z?n0pk?m+1?n0p=k?m+1?(k?n0n?n0z)=1?m+n0n+n0z Wealsohave 1?m+k?m+1=k?2m+2 n0n+n0zk?m+1 Becauseofm=dk=2e, Thus,thenetworkowisofminimumcost. costinallcases. Thecaseofnp+nz?nn<0canbeprovedsimilarly.Thus,thenetworkowisofminimum n0z+n0p?n0nk?2m+20 toconstructaring.applyingthedde-chainalgorithmtotheringwithoutconsideringthe AnexampleisshowninFigure9.Anend-roundedgeisaddedtothechaininFigure4 9 2
10 end-roundedge,theowisshowninfigure9(a).thenumberoftaskstransferredis19.the augmentationisappliedtothisow:np=1;nz=2;nn=5 resultisshowninfigure9(b).thenumberoftaskstransferredisreducedto17. Becausenp+nz?nn<0andthe4thsmallestxi;jis?2,everyxi;jissubtractedby?2.The i=0 4i=1 6i=2 5i=3 (a) i=4 i=5 2i=6 1i=7 i=0 92i=1 74i=2 43i=3 112 Figure9:ExampleforDDE-ring. (b) i=4 42i=5 6 i=6 13i=7 5 n-cube.thealgorithmisshowninfigure10.initerationloftheddealgorithm,subcubeslm isdividedintokpartitionssl+1 WiththeDDE-ringalgorithm,itisnotdiculttocompositeaDDEalgorithmforthek-ary 5.TheDDEMethodforthek-aryn-cube DDE,nodeiwillhavewnitasks.Thetaskexchangestepcanuseeitherthereceive-before-send nodesindierentdimensions.takingatorusandstripthemofalltheend-roundconnections,we orsend-before-receivealgorithm. nodes.thenodesineachringexchangetheirloadandtheneachnodeihaswl+1 Thisalgorithmcanbeappliedtothen-dimensionaltorus,whichallowsdierentnumberof km+bwherem=0;1;:::;kl?1andb=0;1;:::;k?1.slmhaskn?l getamesh.thisalgorithmcanbeappliedtothemeshbyperformingthedde-chainalgorithm itasks.executing ofthek-aryn-cube. insteadofdde-ringineachstep. ThefollowingtheoremshowsthattheloaddierenceofDDEisboundedbyn,thedimension 10
11 DDEfork-aryn-cube Assumeak-aryn-cubeS0,thenumberofnodesiskn,andnodeihasw0itasks. forl=0ton?1 applythedde-ringalgorithmtokn?1ringsinthelthdimensionindependently, whereeveryringhasknodes(a0;a1;:::;al;:::;ak?1)andal=0;1;:::;k?1 exchangetasksaccordingtotheow eachnodeupdatesitsweightwl+1 i=wli+xi?1;i?xi;i+1 Figure10:TheDDEalgorithmforthek-aryn-cube. Theorem1:AfterexecutionofDDE,theloaddierence D=max(wni)?min(wni) isboundedbyn. Proof:InthelthstepofDDE,ak-ary(n?l)-cubeispartitionedintokk-ary(n?l?1)-cubes. Thedierenceofthenumberoftasksbetweentwopartitionsismaximalwhenineachringevery nodeinrstpartition,saysl+1 km,hasonemoretaskthanthatpossessedbythenodeintheother partitions,sl+1 Xkm+b,whereb=1;2;:::;k?1.Thus j2sl+1 kmwl+1 j=x j2sl+1 km+k?1wl+1 j+jsl+1 kmj=1 k?1(x j2slmwlj?x j2sl+1 kmwl+1 j)+kn?l?1 wherejsl+1 kmjdenotesthenumberofnodesinsubcubesl+1 kmwhichiskn?l?1.therefore, X j2sl+1 kmwl+1 j=1kx j2slmwlj+(k?1)kn?l?2 Similarly,X j2sl+1 km+k?1wl+1 j=x j2sl+1 kmwl+1 j?jsl+1 kmj=(x j2slmwlj?(k?1)x j2sl+1 km+k?1wl+1 j)?kn?l?1 Therefore, X j2sl+1 km+k?1wl+1 j=1kx j2slmwlj?kn?l?2 Let Almax=max 0m<klX j2slmwlj 11
12 and Whenl=0, A0max=A0min=X Almin=min j2s0w0j=x 0m<klX j2slmwlj: Similarly, wheretisthetotalnumberoftasks.thus, Almax=(T1kAl?1 max+(k?1)kn?l?1otherwise 0j<knw0j=T Thesolutiontotheaboverecurrenceisgivenby Almin=( 1kAl?1 min?kn?l?1otherwise T ifl=0 ifl=0 (1) Almax=Tkl+(k?1)lkn?l?1 (2) Itclearlysatises(1)and(2)forthebasis,l=0.If(3)satises(1)forl=m,then Am+1 max=t km+1+(m+1)(k?1)kn?(m+1)?1=1k(t Almin=Tkl?lkn?l?1 km+(k?1)mkn?m?1)+(k?1)kn?(m+1)?1 (4) (3) wheneverl0.similarly,itcanbeshownthat(4)satises(2)wheneverl0. Therefore,itsatises(1)forl=m+1.Thus,byinductiononlwehaveshownthat(3)satises(1) Letl=n Anmax=max =1kA(m+1)?1 max+(k?1)kn?(m+1)?1 atmostbyn. BecauseD=Anmax?Anmin=(k?1 Anmin=min k+1k)n=n,thenumberoftasksinanytwoprocessorsdiers 0j<knwnj=Tkn+k?1 0j<knwnj=Tkn?1kn; kn isshowninfigure11(c).themaximumloaddierenceis2. thatthedde-ringalgorithmappliestoeachringintherstdimension.then,dde-ringapplies toeachringintheseconddimension,asshowninfigure11(b).theresultantloaddistribution AnexampleisshowninFigure11.Thisisa4-ary2-cube(i.e.,torus44).Figure11(a)shows 2 12
13 (a) (b) Inthissection,wecompareperformanceofGDEandDDE.Weconsideratestsetofload Figure11:ExampleforDDE(4-ary2-cube). 6.ExperimentalResults (c) speciedvalue.inthissimulationexperiment,theaveragenumberoftasks(averageweight)per distributions,inwhichtheloadateachprocessorisrandomlyselectedwiththemeanequaltoa torus,an8883d-mesh,anda d-torus.forthesenetworks,theoptimalvalueof forgdeis0.723[10]. processoris1,000.eachresultistheaverageof100testcases.wetestedan88mesh,a1616 thanthatofdde. showsitsaverageindierentnetworks.here,theloaddierenceofgdeisfourtosixtimeslarger byn,whereasthatofgdeisboundedbyn(k?1)forthemeshandnk=2forthetorus.figure12 First,wecompareloadimbalanceofGDEandDDE.TheloaddierenceofDDEisbounded 13
14 16 GDE 14 thenumberofsweepssfordierentnetworks.thevalueofsisproportionaltok[10].moreover, 12 sincreaseswiththeaverageweight.tableishowstherelationshipbetweenthenumberofsweeps andtheaveragenumberoftasks,measuredonan88mesh. DDEcompletesloadbalancinginonesweepbutGDEneedsmanysweeps.Figure13shows Figure12:Loaddierence x8 mesh 8x8x8 mesh 16x16 torus 16x16x16 torus 12 GDE 10 DDE 8 ancing.therefore,thetotalnumberofcommunicationsofssweepsare3sc=6sn.fordde, sweephasciterations,wherecisthenumberofcolors.forevennumberofk,c=2n.each iterationhasthreecommunications,twoforexchangingloadinformationandoneforloadbal- 6 therearekcommunicationstepsineachdimensionforcollectionandbroadcastingofloadinformation.loadbalancingneedsatmostk?1andk=2communicationstepsforthemeshandthe torus,respectively.therefore,2knor32kncommunicationstepsintotalarerequired.ddecan Next,wecomparethenumberofcommunicationstepsofGDEandDDE.ForGDE,each Figure13:Thenumberofsweeps. 4 2 reducethenumberofcommunicationstepssignicantly.theanalysishasbeenconrmedbythe 0 experiment,asshowninfigure14. 8x8 mesh 8x8x8 mesh 16x16 16x16x16 14
15 TableI:TheRelationshipBetweentheNumberofSweepsandtheAverageWeight AverageNumberofSweeps AverageNumberofTasks ,0003,00010,000 municationcostisdenedasthethetotalnumbersoftaskstransferreddividedbythetotal numberoftasks: Figure15showsthenormalizedcommunicationcostofGDEandDDE.Thenormalizedcom- Figure14:Thenumberofcommunicationsteps arenotmigratedtoothernodes.ddekeeps20%to50%moretasksinlocal. isabout50%largerthanthatofdde.itisduetothefactthatgdetransferstasksunnecessarily. Finally,DDEhasbetterlocalitythanGDE.Figure16showsthepercentageoflocaltasksthat whereejisthenumberoftaskstransmittedthroughtheedgej.thecommunicationcostofgde Piwi; Pjej x8 mesh 8x8x8 mesh 16x16 torus 16x16x16 torus GDE DDE
16 Figure15:Normalizedcommunicationcost. 8x8 mesh 8x8x8 mesh 16x16 torus 16x16x16 torus GDE DDE 60% 50% 40% 30% 20% 10% 0% Figure16:Thepercentageoflocaltasks. 16 8x8 mesh 8x8x8 mesh 16x16 torus 16x16x16 torus GDE DDE
17 n-cube,ddeisfaster,balancestheloadwell,reducescommunications,andkeepsbetterlocality. tothek-aryn-cube.comparedtothegdealgorithm,whichalsoextendeddemtothek-ary Thispaperproposedadirectmethodforloadbalancing.ItextendedtheDEMalgorithm 7.Conclusion themeshwalkingalgorithm[8].however,dderetainsitssimplicityofimplementationandcan deliverasatisedperformanceatthesametime. References DDEcanbefurtherimprovedforamorebalancedloadandlesscommunicationsbyextending [1]G.Cybenko.Dynamicloadbalancingfordistributedmemorymultiprocessors.J.ofParallel [2]M.Barnettetal.Broadcastingonmesheswithwormholerouting.TechnicalReportTR-93- [3]S.H.Hosseini,B.Litow,M.Malkawi,J.McPherson,andK.Vairavan.Analysisofagraph Distrib.Comput.,7:279{301,1989. coloringbaseddistributedloadbalancingalgorithm.journalofparallelanddistributed 24,Univ.TexasatAustin,1993. [6]W.ShuandM.Y.Wu.Runtimeparallelschedulingfordistributedmemorycomputers.In [4]E.L.Lawler.CombinatorialOptimization:NetworksandMatroids.Holt,Rinehartand [5]S.Ranka,Y.Won,andS.Sahni.Programmingahypercubemulticomputer.IEEESoftware, Int'lConf.onParallelProcessing,pagesII.143{150,August1995. pages69{77,september1988. Winston,1976. Computing,10:160{166,1990. [7]MarcWillebeek-LeMairandAnthonyP.Reeves.Strategiesfordynamicloadbalancing onhighlyparallelcomputers.ieeetrans.parallelanddistributedsystem,9(4):979{993, [10]C.Z.XuandF.C.M.Lau.Thegeneralizeddimensionexchangemethodforloadbalancing [8]M.Y.WuandW.Shu.High-performanceincrementalschedulingonmassivelyparallelcomputers aglobalapproach.insupercomputing'95,december1995. September1993. [9]C.Z.XuandF.C.M.Lau.Analysisofthegeneralizeddimensionexchangemethodfor December1992. January1995. ink-aryn-cubesandvariants.journalofparallelanddistributedcomputing,24(1):72{85, dynamicloadbalancing.journalofparallelanddistributedcomputing,16(4):385{393, 17
Graphic Algorithms and the Demographic Variations
DDE:AModiedDimensionExchangeMethod forloadbalancingink-aryn-cubes StateUniversityofNewYorkatBualo DepartmentofComputerScience Min-YouWuandWeiShu algorithmforthehypercubestructure.ithasbeengeneralizedtok-aryn-cubes.however,the
Mail for OS X Medical School IMAP & Exchange Email Setup Guide
MailforOSX MedicalSchoolIMAP&ExchangeEmail SetupGuide TherearetwowaystosetupyourMedicalSchoolemailaccountusingMail.One wayisconnectingusingexchangeandtheotherisbyusingimap.thisguidewill showyouhowtosetupmailusingthesetwomethods.forbestresults,we
Components: Interconnect Page 1 of 18
Components: Interconnect Page 1 of 18 PE to PE interconnect: The most expensive supercomputer component Possible implementations: FULL INTERCONNECTION: The ideal Usually not attainable Each PE has a direct
Nearestneighboralgorithmsforloadbalancingin ChengzhongXu parallelcomputers DepartmentofElectricalandComputerEngg. WayneStateUniversity,Detroit,MI48202 BurkhardMonien,ReinhardLuling [email protected]
Factoring Special Polynomials
6.6 Factoring Special Polynomials 6.6 OBJECTIVES 1. Factor the difference of two squares 2. Factor the sum or difference of two cubes In this section, we will look at several special polynomials. These
1 Shapes of Cubic Functions
MA 1165 - Lecture 05 1 1/26/09 1 Shapes of Cubic Functions A cubic function (a.k.a. a third-degree polynomial function) is one that can be written in the form f(x) = ax 3 + bx 2 + cx + d. (1) Quadratic
Topological Properties
Advanced Computer Architecture Topological Properties Routing Distance: Number of links on route Node degree: Number of channels per node Network diameter: Longest minimum routing distance between any
x 3 x 4 x 2 f -7.0-9.5 4.3-4.2-9.0-8.7-3.5-5.7-8.0 3.4-6.9-8.7 9.8 3.8 5.1-2.7-6.5 0.8-3.5 7.1 6.2 9.9 7.7 0.1-7.4 Site A Site B Site C
DistributedMultivariateRegressionUsing Wavelet-basedCollectiveDataMining SchoolofElectricalEngineeringandComputerScience DarylE.HershbergerandHillolKargupta SchoolofElectricalEngineeringandComputerScienceTechnicalReportEECS99-002
Figure2:Themixtureoffactoranalysisgenerativemodel. j;j z
TheEMAlgorithmforMixturesofFactorAnalyzers DepartmentofComputerScience ZoubinGhahramani GeoreyE.Hinton May21,1996(revisedFeb27,1997) TechnicalReportCRG-TR-96-1 Email:[email protected] Toronto,CanadaM5S1A4
... Schema Integration
DataIntegrationTechniquesbasedon MichaelGertz DataQualityAspects DepartmentofComputerScience UniversityofCalifornia,Davis IngoSchmitt [email protected] Davis,CA95616,USA OneShieldsAvenue Otto-von-Guericke-UniversitatMagdeburg
{apolin},{mcampos}@ieee.org
{apolin},{mcampos}@ieee.org x( ) 2 x( ) x( ) = ( ) x = [ ( ) x ı x + ( ) y ( ) y ( ) z ı y + ( ) z ] T ı z 2 x ( ) = 2 ( ) x + 2 ( ) 2 y + 2 ( ) 2 z 2 2 E = 1 2 E c 2 t 2 s(x,t) 2 s x + 2 s 2 y + 2
Chapter 2. Multiprocessors Interconnection Networks
Chapter 2 Multiprocessors Interconnection Networks 2.1 Taxonomy Interconnection Network Static Dynamic 1-D 2-D HC Bus-based Switch-based Single Multiple SS MS Crossbar 2.2 Bus-Based Dynamic Single Bus
Actuarial mathematics 2
Actuarial mathematics 2 Life insurance contracts Edward Furman Department of Mathematics and Statistics York University January 3, 212 Edward Furman Actuarial mathematics MATH 328 1 / 45 Definition.1 (Life
Introduction to the Finite Element Method (FEM)
Introduction to the Finite Element Method (FEM) ecture First and Second Order One Dimensional Shape Functions Dr. J. Dean Discretisation Consider the temperature distribution along the one-dimensional
Factoring Trinomials: The ac Method
6.7 Factoring Trinomials: The ac Method 6.7 OBJECTIVES 1. Use the ac test to determine whether a trinomial is factorable over the integers 2. Use the results of the ac test to factor a trinomial 3. For
Interconnection Network
Interconnection Network Recap: Generic Parallel Architecture A generic modern multiprocessor Network Mem Communication assist (CA) $ P Node: processor(s), memory system, plus communication assist Network
Themethodofmovingcurvesandmovingsurfacesisanew,eectivetoolfor Abstract
OnaRelationshipbetweentheMovingLineand MovingConicCoecientMatrices DepartmentofComputerScience Houston,Texas77005 [email protected] RiceUniversity MingZhang DepartmentofInformationSystemsandComputerScience
Sleeve Yokes / Kayıcı Çatallar
Sleeve Yokes / Kayıcı Çatallar 19,00 126,00 49,70 Spline / Kanal SE 1 1/16 x 16 ngle / çı 25 0 Serie 0400 MS NO SY 01 23,80 149,00 64,20 Spline / Kanal SE 1 x 10 ngle / çı 20 0 Serie 1140 MS NO SY 02 23,80
"#$%&'((&)!*+,-./ 0+1$23!4-+5#.-)!!
"#$%&'((&)*+,-./ 0+1$234-+5#.-) 6(75()+1-,+,-.+289(2-:7$5;"#$%&'((&) JoeWoodard,President CreativeFinancialSoftware,Inc. AdvancedCertifiedQuickBooksProAdvisor Member"IntuitSolutionProviderProgram Intuit
Using the ac Method to Factor
4.6 Using the ac Method to Factor 4.6 OBJECTIVES 1. Use the ac test to determine factorability 2. Use the results of the ac test 3. Completely factor a trinomial In Sections 4.2 and 4.3 we used the trial-and-error
Introduction: Overview of Kernel Methods
Introduction: Overview of Kernel Methods Statistical Data Analysis with Positive Definite Kernels Kenji Fukumizu Institute of Statistical Mathematics, ROIS Department of Statistical Science, Graduate University
Technical specification bucket JET 53 P
Technical specification bucket Jokey Euro Tainer Profi D1: 227,0 mm D2: 21 D3: 192,0 mm H: 18 basic data volume to rim: 5,70 l handle types tolerances 143,0 g metal handle 21,0 g : plastic handle 10,0
Factoring Trinomials of the Form x 2 bx c
4.2 Factoring Trinomials of the Form x 2 bx c 4.2 OBJECTIVES 1. Factor a trinomial of the form x 2 bx c 2. Factor a trinomial containing a common factor NOTE The process used to factor here is frequently
System Interconnect Architectures. Goals and Analysis. Network Properties and Routing. Terminology - 2. Terminology - 1
System Interconnect Architectures CSCI 8150 Advanced Computer Architecture Hwang, Chapter 2 Program and Network Properties 2.4 System Interconnect Architectures Direct networks for static connections Indirect
Change Discussion Guide
Example: Budget Coordinator About This Guide What is a Change Discussion Guide? The Change Discussion Guide describes process changes that affect the way you will
AClassofLinearAlgorithmstoProcessSetsofSegments GonzaloNavarroRicardoBaeza-Yates DepartmentofComputerScience fgnavarro,[email protected] BlancoEncalada2120 UniversityofChile Santiago-Chile currentsolutionstooperatesegmentsfocusonsingleoperations(e.g.insertionorsearching),
Section 4.5 Exponential and Logarithmic Equations
Section 4.5 Exponential and Logarithmic Equations Exponential Equations An exponential equation is one in which the variable occurs in the exponent. EXAMPLE: Solve the equation x = 7. Solution 1: We have
Parallel Programming
Parallel Programming Parallel Architectures Diego Fabregat-Traver and Prof. Paolo Bientinesi HPAC, RWTH Aachen [email protected] WS15/16 Parallel Architectures Acknowledgements Prof. Felix
Self-piercing riveting
Self-piercing riveting Andre Stühmeyer, CAD-FEM GmbH Germany Self-piercing riveting Overview The self piercing riveting process FE analysis of the joining process Large deformation Material failure 2D
MERITOR. FEKO No : 3164 FH OEM No : Application/Kullanım yeri : ELSA 2 Description/Açıklama : Triangle Seal/ Üçgen Keçe Dimensions/Ölçü mm : 52x6,70
FEKO No 3480 FE-Sİ-GR - Toz Lastiği 32,30x97,30x12,50 FEKO No 3190 FE-Sİ-GR MCK1139 ELSA 1--- 29x51,50x25 FEKO No 3164 FH yeri Triangle Seal/ Üçgen Keçe 52x6,70 FEKO No 2877 FH yeri Actuator Seal/Fren
ASocialMechanismofReputationManagement inelectroniccommunities 446EGRC,1010MainCampusDrive BinYuandMunindarP.Singh? DepartmentofComputerScience NorthCarolinaStateUniversity fbyu,[email protected] Raleigh,NC27695-7534,USA
A. Factoring out the Greatest Common Factor.
DETAILED SOLUTIONS AND CONCEPTS - FACTORING POLYNOMIAL EXPRESSIONS Prepared by Ingrid Stewart, Ph.D., College of Southern Nevada Please Send Questions and Comments to [email protected]. Thank you!
NP-completeproblemstractable Copyingquantumcomputermakes MikaHirvensalo February1998 ISBN952-12-0158-4 TurkuCentreforComputerScience ISSN1239-1891 TUCSTechnicalReportNo161 superpositions,weshowthatnp-completeproblemscanbesolvedprobabilisticallyinpolynomialtime.wealsoproposetwomethodsthatcould
Manual for SOA Exam MLC.
Chapter 5 Life annuities Extract from: Arcones Manual for the SOA Exam MLC Fall 2009 Edition available at http://wwwactexmadrivercom/ 1/70 Due n year deferred annuity Definition 1 A due n year deferred
Lecture 18: Interconnection Networks. CMU 15-418: Parallel Computer Architecture and Programming (Spring 2012)
Lecture 18: Interconnection Networks CMU 15-418: Parallel Computer Architecture and Programming (Spring 2012) Announcements Project deadlines: - Mon, April 2: project proposal: 1-2 page writeup - Fri,
Section 3.2 Polynomial Functions and Their Graphs
Section 3.2 Polynomial Functions and Their Graphs EXAMPLES: P(x) = 3, Q(x) = 4x 7, R(x) = x 2 +x, S(x) = 2x 3 6x 2 10 QUESTION: Which of the following are polynomial functions? (a) f(x) = x 3 +2x+4 (b)
Hyper Node Torus: A New Interconnection Network for High Speed Packet Processors
2011 International Symposium on Computer Networks and Distributed Systems (CNDS), February 23-24, 2011 Hyper Node Torus: A New Interconnection Network for High Speed Packet Processors Atefeh Khosravi,
Advanced Microeconomics
Advanced Microeconomics Ordinal preference theory Harald Wiese University of Leipzig Harald Wiese (University of Leipzig) Advanced Microeconomics 1 / 68 Part A. Basic decision and preference theory 1 Decisions
Attention windows of second level fixations. Input image. Attention window of first level fixation
HandSegmentationUsingLearning-BasedPredictionand VericationforHandSignRecognition DepartmentofComputerScience YuntaoCuiandJohnJ.Weng mentationschemeusingattentionimagesfrommultiple Thispaperpresentsaprediction-and-vericationseg-
Question 1a of 14 ( 2 Identifying the roots of a polynomial and their importance 91008 )
Quiz: Factoring by Graphing Question 1a of 14 ( 2 Identifying the roots of a polynomial and their importance 91008 ) (x-3)(x-6), (x-6)(x-3), (1x-3)(1x-6), (1x-6)(1x-3), (x-3)*(x-6), (x-6)*(x-3), (1x- 3)*(1x-6),
Problem 1: Computation of Reactions. Problem 2: Computation of Reactions. Problem 3: Computation of Reactions
Problem 1: Computation of Reactions Problem 2: Computation of Reactions Problem 3: Computation of Reactions Problem 4: Computation of forces and moments Problem 5: Bending Moment and Shear force Problem
Polynomial Expression
DETAILED SOLUTIONS AND CONCEPTS - POLYNOMIAL EXPRESSIONS Prepared by Ingrid Stewart, Ph.D., College of Southern Nevada Please Send Questions and Comments to [email protected]. Thank you! PLEASE NOTE
Computational Physics
Computational Physics Sheet 6, Computational Physics Course 17105 Professor: H. Ruhl, Exercises: N. Moschüring and N. Elkina Discussion of Solutions: Dec 03, 01, Room A49 Problem 1: Yee solver for Schrödinger
How To Lead In A Contract
STANDARD OPENING LEADS (WHICH CARD) There are two main aspects to deciding the opening lead; which suit to lead, and which card in that suit. Here we shall only consider which card to lead in a specific
Real-TimeVericationofStatemateDesigns. applicationsraisesthedemandforprovingtheircorrectness.becauseverication
Real-TimeVericationofStatemateDesigns vericationofstatemate1designs.statemateisawidelyuseddesign Abstract.Thispaperpresentsanapproachtowardsreal-time toolforembeddedcontrolunits.theseembeddedcontrolunitsare
DistributedSharedMemorySystems? AdaptiveLoadBalancinginSoftware CompilerandRun-TimeSupportfor SotirisIoannidisandSandhyaDwarkadas fsi,[email protected] DepartmentofComputerScience Rochester,NY14627{0226
From DeLuna Tape #002 05:22:00 to DeLuna Tape #002 05:40:54
DeLuna Tape #002 From DeLuna Tape #002 05:22:00 to DeLuna Tape #002 05:40:54 Diamond Shamrock at night. [NOTE: This walk-through was recorded on December 4, 2004, before investigators interviewed Rene
Mathematics of Life Contingencies MATH 3281
Mathematics of Life Contingencies MATH 3281 Life annuities contracts Edward Furman Department of Mathematics and Statistics York University February 13, 2012 Edward Furman Mathematics of Life Contingencies
Number of objects. 16 32 64 128 256 512 1k 2k 4k 8k 16k 32k 64k 128k256k512k 1m 2m 4m 8m
GarbageCollectionforLargeMemoryJava AndreasKrallandPhilippTomsich Applications InstitutfurComputersprachen,TechnischeUniversitatWien Argentinierstrae8,A{1040Wien,Austria tolarge,data-intensivescienticapplicationsallocatingmemoryinthe
On-Chip Interconnection Networks Low-Power Interconnect
On-Chip Interconnection Networks Low-Power Interconnect William J. Dally Computer Systems Laboratory Stanford University ISLPED August 27, 2007 ISLPED: 1 Aug 27, 2007 Outline Demand for On-Chip Networks
F. P. Beer et al., Meccanica dei solidi, Elementi di scienza delle costruzioni, 5e - isbn 9788838668579, 2014 McGraw-Hill Education (Italy) srl
F. P. Beer et al., eccanica dei solidi, Elementi di scienza delle costruzioni, 5e - isbn 9788888579, 04 cgraw-hill Education (Italy) srl Reactions: Σ = 0: bp = 0 = Pb Σ = 0: ap = 0 = Pa From to B: 0
Recitation #5. Understanding Shear Force and Bending Moment Diagrams
Recitation #5 Understanding Shear Force and Bending Moment Diagrams Shear force and bending moment are examples of interanl forces that are induced in a structure when loads are applied to that structure.
Red Hat Enterprprise Linux - New Offerings SYSTEM OPTIONS
Red Hat Enterprprise Linux - New Offerings SYSTEM OPTIONS PRODUCT CODE Red Hat Enterprise Linux for Servers DESCRIPTION Support Level No No 1 Year 1 Year 3 Year 3 Year Sockets Virtual Guests New Renewal
SSLV105 - Stiffening centrifuges of a beam in rotation
Titre : SSLV105 - Raidissement centrifuge d'une poutre en [...] Date : 19/09/2011 Page : 1/6 SSLV105 - Stiffening centrifuges of a beam in rotation Summarized: Test of Structural mechanics in linear static
Design and Implementation of a P2P Cloud System
Design and Implementation of a P2P Cloud Ozalp Babaoglu, Moreno Marzolla, Michele Tamburini {babaoglu, marzolla}@cs.unibo.it, [email protected] Our Goal Assemble a Cloud out of individual devices E.g.
Deferred Annuities Certain
Deferred Annuities Certain General terminology A deferred annuity is an annuity whose first payment takes place at some predetermined time k + 1 k n a... the present value of a basic deferred annuity-immediate
Networks on Chip. on-chip interconnect: physical. Kees Goossens. Kees Goossens <[email protected]> Eindhoven University of Technology 1
Eindhoven University of Technology 1 Networks on Chip Kees Goossens Kees Goossens Group Electrical Engineering Faculty on-chip interconnect: physical Kees
Aluminum Capacitors Radial Style
Component outlines Obsolete - please refer to: www.vishay.com/doc?25013 FEATURES Polarized Aluminum electrolytic capacitor High C x U product Small dimensions Long lifetime Extended temperature range:
Fast Fourier Transform: Theory and Algorithms
Fast Fourier Transform: Theory and Algorithms Lecture Vladimir Stojanović 6.973 Communication System Design Spring 006 Massachusetts Institute of Technology Discrete Fourier Transform A review Definition
Chapter 12: Multiprocessor Architectures. Lesson 04: Interconnect Networks
Chapter 12: Multiprocessor Architectures Lesson 04: Interconnect Networks Objective To understand different interconnect networks To learn crossbar switch, hypercube, multistage and combining networks
3.2. Solving quadratic equations. Introduction. Prerequisites. Learning Outcomes. Learning Style
Solving quadratic equations 3.2 Introduction A quadratic equation is one which can be written in the form ax 2 + bx + c = 0 where a, b and c are numbers and x is the unknown whose value(s) we wish to find.
Interconnection Networks
CMPT765/408 08-1 Interconnection Networks Qianping Gu 1 Interconnection Networks The note is mainly based on Chapters 1, 2, and 4 of Interconnection Networks, An Engineering Approach by J. Duato, S. Yalamanchili,
Technical Note 3175A Fault finding Cooling circuit
Technical Note 3175A Fault finding Cooling circuit Vehicles concerned: see list on the following page Subsections concerned: Fault finding procedures for various faults affecting the cooling system and
MATH 110 Automotive Worksheet #4
MATH 110 Automotive Worksheet #4 Ratios The math name for a fraction is ratio. It is just a comparison of one quantity with another quantity that is similar. As an automotive technician, you will use ratios
Load balancing in a heterogeneous computer system by self-organizing Kohonen network
Bull. Nov. Comp. Center, Comp. Science, 25 (2006), 69 74 c 2006 NCC Publisher Load balancing in a heterogeneous computer system by self-organizing Kohonen network Mikhail S. Tarkov, Yakov S. Bezrukov Abstract.
PARTICIPANT INFORMATION. The Oxford Access for Students Improving Sleep (OASIS) Study
PARTICIPANT INFORMATION The Oxford Access for Students Improving Sleep (OASIS) Study We would like to invite you to take part in a research study. This webpage should provide you with all the information
Finding New Opportunities with Predictive Analytics. Stephanie Banfield 2013 Seminar for the Appointed Actuary Session 4 (Life)
Finding New Opportunities with Predictive Analytics Stephanie Banfield 2013 Seminar for the Appointed Actuary Session 4 (Life) Agenda First a Story What is Predictive Analytics? Predictive Underwriting
threads threads threads
AHybridMultithreading/Message-PassingApproachforSolving IrregularProblemsonSMPClusters Jan-JanWu InstituteofInformationScience AcademiaSinica Taipei,Taiwan,R.O.C. Chia-LienChiang Nai-WeiLin Dept.ComputerScience
Òàëðåïû, çàêðûòîãî òèïà Turnbuckles, closed type B 1 - B 5
Òàëðåïû, çàêðûòîãî òèïà Turnbuckles, closed type B 1 - B 5 B 1 Hook - hook B 2 Hook - jaw B 3 Jaw - jaw B 4 Hook - spacer B 5 Handwheel B 1 B 2 B 3 / / / 18,0 18,0 18,0 min-max () 950-1350 950-1350 950-1350
A Source Identification Scheme against DDoS Attacks in Cluster Interconnects
A Source Identification Scheme against DDoS Attacks in Cluster Interconnects Manhee Lee* Eun Jung Kim* Cheol Won Lee *Department of Computer Science Texas A&M University College Station, TX-77840 [email protected],
4. Expanding dynamical systems
4.1. Metric definition. 4. Expanding dynamical systems Definition 4.1. Let X be a compact metric space. A map f : X X is said to be expanding if there exist ɛ > 0 and L > 1 such that d(f(x), f(y)) Ld(x,
Support Vector Machines
Support Vector Machines Charlie Frogner 1 MIT 2011 1 Slides mostly stolen from Ryan Rifkin (Google). Plan Regularization derivation of SVMs. Analyzing the SVM problem: optimization, duality. Geometric
Problem set 2, Part 2: Generalized Roy Model 2 Factor, no normality
Problem set 2, Part 2: Generalized Roy Model 2 Factor, no normality After doing this problem set you should be able to figure out how to include more factors (so you make the model more flexible) get rid
Sensitivity Analysis of Risk Measures for Discrete Distributions
Sensitivity Analysis of Risk Measures for Discrete Distributions Jean-Paul Laurent 1 Preliminary Version August 2003 Abstract We consider the computation of quantiles and spectral risk measures for discrete
Math 370, Spring 2008 Prof. A.J. Hildebrand. Practice Test 2
Math 370, Spring 2008 Prof. A.J. Hildebrand Practice Test 2 About this test. This is a practice test made up of a random collection of 15 problems from past Course 1/P actuarial exams. Most of the problems
Procedural Animation. An introduction
Procedural Animation An introduction What is procedural? Procedural animation is an approach (i.e. a way) to create animation. It is not a style of animation. But the animation created by procedural approach
Cray Gemini Interconnect. Technical University of Munich Parallel Programming Class of SS14 Denys Sobchyshak
Cray Gemini Interconnect Technical University of Munich Parallel Programming Class of SS14 Denys Sobchyshak Outline 1. Introduction 2. Overview 3. Architecture 4. Gemini Blocks 5. FMA & BTA 6. Fault tolerance
PART I: A STANDARD ANALYSIS OF FACTOR MOBILITY
1 PART I: A STANDARD ANALYSIS OF FACTOR MOBILITY CHAPTER 1: BENEFICIAL vs. DISTORTIONARY MOBILITY OF FACTORS OF PRODUCTION Introduction Classical economic setups suggest that factors of production move,
