SOAPUI压力测试的参数配置.docx

SOAPUI压力测试的参数配置.docx

《SOAPUI压力测试的参数配置.docx》由会员分享，可在线阅读，更多相关《SOAPUI压力测试的参数配置.docx（10页珍藏版）》请在冰豆网上搜索。

SOAPUI压力测试的参数配置

SOAPUI压力测试的参数配置

ThedifferentLoadStrategiesavailableinsoapUIandsoapUIProallowyoutosimulatevarioustypesofloadovertime,enablingyoueasilytesttheperformanceofyourtargetservicesunderanumberofconditions.SincesoapUIalsoallowsyoutorunmultipleLoadTestssimultaneously （seeanexamplefurtherdown）,acombinationofLoadTestscanbeusedtofurtherassertthebehaviourofyourservices.SelectthedesiredStrategyforyourLoadTestfromtheStrategyToolbarintheLoadTestWindow:

LetshavealookatthedifferentLoadStrategiesavailableandseehowtheycanbeusedtododifferenttypesofLoad/Performancetests.

1. SimpleStrategy-Baseline,LoadandSoakTesting

TheSimpleStrategyrunsthespecifiednumberofthreadswiththespecifieddelaybetweeneachruntosimulatea"breathingspace"fortheserver.Forexampleifyouwanttorunafunctionaltestwith10threadswith10secondsdelay,setThreadsto10,delayto10000andrandomtohowmuchofthedelayyouwanttorandomize（iesettingitto0.5willresultindelaysbetween5and10seconds）.WhencreatinganewLoadTestthisisthedefaultstrategyandsetatarelativelylowload（5threadswith1000msdelay）.

TheSimpleStrategyisperfectforBaselinetesting.Useittoassertthethebasicperformanceofyourserviceandvalidatethattherearenothreadingorresourcelockingissues.Rampupthenumberofthreadswhenyouwantdomoreelaborateloadtestingorusethestrategyforlong-runningsoaktests.

Sinceitisn'tmeanttobringyourservicestotheirknees,asetuplikethiscanbeusedforcontinuousload-testingtoensurethatyourserviceperformsasexpectedundermoderateload;setupabaselinetestwithnorandomizationofthedelay,addLoadTestAssertionsthatactasasafetynetforunexpectedresultsandautomateitsexecutionwiththecommand-lineLoadTestrunnerormavenplugins.

2. FixedRateStrategy–Simplewithatwist

OnethingthattheSimpleStrategydoesnotdoisguaranteeanumberofexecutionswithinacertaintime,forexampleifyoumightwanttostartyourTestCase10timeseachsecondnomatterhowlongittakestoexecute.UsingtheSimpleStrategyyoucouldsetup10ThreadsandadelaycompensatingfortheaveragegapbetweentheendoftheTestCaseandthestartofthenextsecond,butthiswouldbehighlyunreliableovertime.TheFixed-Ratestrategyshouldbeusedinstead;settherateasdesired（10inourcase）andoffyougo;thestrategywillautomaticallystarttherequirednumberthreadsforthissettingattemptingtomaintaintheconfiguredvalue.

Ashintedintheheadline,therearesometwistshere:

whatifourTestCasetakesmorethanonesecondtoexecute?

TomaintaintheconfiguredTPSvalue,thestrategywillinternallystartnewthreadstocompensateforthis;afterawhileyoumighthavemanymorethan10threadsrunningduetothefactthattheoriginaloneshadnotfinishedwithinthesetrate.Andnotsurprisinglythiscouldcausethetargetservicetogetevenslower,resultinginmoreandmorethreadsbeingstartedto"keepup"withtheconfiguredTPSvalue.Asyouprobablyguessedthe"MaxThreads"settingishertopreventsoapUIfromoverloading（bothitselfandthetargetservices）  inthissituation,specifyingavalueherewillputalimitonthemaximumnumberofthreadsthesoapUIwillbeallowedtostarttomaintaintheconfiguredTPS,ifreachedtheexistingthreadswillhavetofinishbeforesoapUIwillstartanynewones.

The"RequestLevel"settingwillattempttomaintaintheTPSnotontheTestCaseexecutionlevelbutontherequestlevelinstead,forexampleifyouhaveadata-drivenLoadTestoraTestCasewithmanyrequests,youwanttheTPSsettingtoapplynotontheexecutionleveloftheentireTestCasebutontherequestlevel.

Inanycase,theFixedRatestrategyisusefulforbaseline,loadandsoak-testingifyoudon’trunintothe"ThreadCongestion"problemdescribedabove.Ontheotherhand,youmightprovokethecongestion（maybeevenincombinationwithanotherLoadTest）toseehowyourserviceshandlethisorhowtheyrecoverafterthecongestionhasbeenhandled.

3. VariableLoadStrategies

Thereareseveralstrategiesthatcanbeusedtovaryload（thenumberofthreads）overtime,eachsimulatingadifferentkindofbehavior.Theycanbeusefulforrecoveryandstresstesting,butjustaswellforbaselinetesting,eitherontheirownorincombinationwithotherstrategies.Let'shaveaquicklook:

1.Variancestrategy –thisvariesthenumberofthreadsovertimeina“sawtooth”manorasconfigured;settheIntervaltothedesiredvalue  andtheVariancetothehowmuchthenumberofthreadsshoulddecreaseandincrease.Forexampleifwestartwith20threads,setintervalto60andVarianceto0.8,thenumberofthreadswillincreasefrom20to36withinthefirst15seconds,thendecreasebackto20andcontinuedownto4threadsafter45seconds,andfinallygobackuptotheinitialvalueafter60seconds.IntheStatisticsDiagramswecanfollowthisvarianceeasily:

2.BurstStrategy -  thisstrategyisspecificallydesignedforRecoveryTestingandtakesvariancetoitsextreme;itdoesnothingfortheconfiguredBurstDelay,thenrunstheconfigurednumberofthreadsforthe“BurstDuration”  andgoesbacktosleep.Hereyoucould（andshould!

）setthenumberofthreadstoahighvalue（20+）tosimulateanonslaughtoftrafficduringashortinterval,thenmeasuretherecoveryofyoursystemwithastandardbase-lineLoadTestcontainingbasicperformance-relatedassertions.Letstrythiswithaburstdelayanddurationof10secondsfor60seconds;

Herecanseetheburstsofactivityinthediagram,pleasealsonotethattheresolutionhasbeenchangedto250ms（fromthedefault"data"value）,otherwisewenothavehadanydiagramupdatesduringthe"sleeping"periodsoftheexecution（sincenodatawouldhavebeencollected）.

3.The ThreadStrategy letsyoulinearly changethenumberofthreadsfromoneleveltoanotherovertherunoftheLoadTest.It’smainpurposeisasameanstoidentifyatwhichlevelcertainstatisticschangeoreventsoccur,forexampletofindatwhichThreadCountthemaximumTPSorBPScanbeachievedortofindatwhichThreadCountfunctionaltestingerrorsstartoccurring. Setthestartandendthreadvalues（forexample5to50）andsetthedurationtoarelativelylongduration（I  useatleast30secondsperthreadvalue,inthisexamplethatwouldbe1350seconds）togetaccuratemeasurements（moreonthisbelow）.

4.GridStrategy –thisstrategyallowstospecificallyconfiguretherelativechangeinnumberofthreadsovertime,itsmainuseforthisismoreadvancedscenarioandrecoverytesting,whereyouneedtoseetheservicesbehaviorundervaryingloadsandloadchanges.Forexampleletssayyouwanttorunfor60secondswith10,20,10,40,10threads.ConfigureyourLoadTesttostartwith10threadsandthenenterthefollowingvaluesintheGrid:

BothvaluesarestoredrelativetothedurationandactualThreadCountoftheLoadTest;ifyouchangethese,thecorrespondingGridStrategyvalueswillberecalculated. Runningthetestshowsthefollowingoutput:

5.ScriptStrategy –thescriptstrategyistheultimatecustomizationpossibility;thescriptyouspecifyiscalledregularly （The"StrategyInterval"settingintheLoadTestOptionsdialog）andshouldreturnthedesirednumberofthreadsatthatcurrenttime.Returningavalueotherthanthecurrentonewillstartorstopthreadstoadjustforthechange.Thisallowsforanykindofvarianceofthenumberofthreads,forexamplethefollowingscriptrandomizesthenumberofthreadsbetween5and15.

Runningthiswiththestrategyintervalsetto5000thenumberofthreadswillchangeevery5seconds:

Thepossibilitieshereareendless.

4. StatisticsCalculationandThreadCountChanges

Manyofthesestrategieswillchangethenumberofthreadswhichhasanimportantimpactonthestatisticscalculationthatyouneedtobeawareof;whenthenumberofthreadschanges,thiswillusuallychangetheresponsetimesofthetargetservices,resultinginachangeinavg,tps,etc. butsincetheLoadTesthasalreadyrunatapreviousnumberofthreadstheresultsforthoserunswillskewtheresultforthenewThreadCount.

Forexampleletssayyouhavebeenrunningat5threadsandgotandaverageto500ms.UsingtheThreadStrategyyouincreasethenumberofthreadsgradually;whenrunning6threadstheaverageincreasesto600msbutsincethe“old”valuescollected  for5threadsarestillthere,thesewillintotalresultinaloweraverage.Therearetwoeasywaystoworkaroundthis;selectthe“ResetStatisticsonThreadCountchange”valueintheLoadTestOptionsdialog,ormanuallyresetthestatisticswiththecorrespondingbuttonintheLoadTesttoolbar;ineithercaseoldstatisticswillbediscarded.ToseethisinactionletsdoaThreadCountStrategytestfrom10to20threadsover300seconds（30secondsperthread）,belowyoucanseeresultsbothwiththissettinguncheckedandthenchecked;

Inthelatteryouseethe“jumps”instatisticseachtimetheyareresetwhenthenumberofthreadschanges,graduallylevelingouttoanewvalue.ThefinalTPScalculatedat20threadsdiffersabout10%betweenthesetwo,showinghowthelowerresults“impact”thehigherones.

5. RunningMultipleLoadTestsSimultaneously

Ok,letshaveaquicklookatthis;we'llcreateonebaselinetestwiththesimplestrategyandalownumberofthreads,andatthesametimerunaburststrategytoseehowthebaselinetestperformance"recovers"aftertheburst;

Hereyoucanseethesimplestrategy（bottomdiagram）recoveringgradua