改進(jìn)的蟻群算法在網(wǎng)格任務(wù)調(diào)度中的研究.pdf

1、大連交通大學(xué)碩士學(xué)位論文改進(jìn)的蟻群算法在網(wǎng)格任務(wù)調(diào)度中的研究姓名：符曉青申請(qǐng)學(xué)位級(jí)別：碩士專(zhuān)業(yè)：計(jì)算機(jī)應(yīng)用技術(shù)指導(dǎo)教師：馬海波20091222。人迕變通久學(xué)1：學(xué)碩十學(xué)何論文AbstractJobscheduleisakeyresearchsubjectincomputinggridIngridenvironments，fromtasksubmissiontoresultprocessing，alleventsabouttasksare

2、underthecontroloftaskmanagementHowtoschedulejobtomeettheusers’needsisaquitechallengingquestionbecausegridpronouncesthecharacteristicsoflargescale，isomerization，dynamics，distributionandautonomy，ectButAntAlgorithmisaeffici

3、entalgorithmtosolvethiskindofproblemThispaperoffersaloadbalancingalgorithmforquickAntAlgorithmonbasisofbreadthfirstsearchandthatisbasedontheresultsofresearchingforthe鰣djobschedulealgorithmwiththecharacteristicsofgridFirs

4、t，analysisthegridcharacteristics，gridsystemstructure，gridresourcesscheduleperiod，offeringantalgorithmthroughapproachtosolvenewintelligentcalculationproblemsandsimulatingnaturalecologicalmechanism，itisquitesuitableforgrid

5、jobschedulewithitsparallelismandextensibilityButthismethodisnotgoodenoughatinefficientandloadimbalance，andwehaveimprovedthemalreadySothattobalancethefast—balanceofthetwomachines，andthenthroughrepeatediterationstofastbala

6、ncethetaskoneachmachineingridsystemFinally，theuseof鰣dschedulingsimulationtoolkitGridSim，cardedoutalargenumberofsimulationexperimentsTheresultsshowedthat：FastBalancealgorithmhasgoodperformance，andCanreceivebetterMin—Min，M