面向服务的云数据挖掘引擎的研究术.pdfVIP

ISSN 1673—9418 CODEN JKYTA8 E—mail：fcst@vip．163．com JournalofFrontiersofComputerScienceandTechnology http：／／www．ceaj．org 1673--9418／2012／06(01)-·0046-·12 Te1：+86．10 DOI：10．37780．issn．1673-9418．2012．01．003 面向服务的云数据挖掘引擎的研究 余永红 ，2+，向晓军 ，高 阳 ，商 琳 ，杨育彬 I．南京大学 计算机软件新技术国家重点实验室，南京 210093 2．南京邮电大学 通达学院，南京 210003 ResearchonService．OrientedDataM iningEngineBasedonCloudComputing YUYonghong，XIANGXiaojun，GAOYang，SHANGLin!，YANGYubin 1．StateKeyLaboratoryforNovelSoftwareTechnology，NanjingUniversity，Nanjing210093，China 2．CollegeofTongda，NanjingUniversityofPostsandTelecommunications，Nanjing210003，China +Correspondingauthor：E—mail：yuyh．nju@gmail．com YUYonghong，XIANGXiaojun，GAOYang，eta1．Researchonservice·orienteddataminingenginebasedon cloudcomputing．JournalofFrontiersofComputerScienceandTechnology,2012，6(1)：46_-57． Abstract：Thescalabilityofdataminingalgorihtmsisrestrictedwhendealingwiht lrage—scaledata．Th ereraesig— nificantdifferencesinawiderangeofapplicationraeasandrequirementsofrknowledgediscovery process．Itis fundamentaltoprovideeffectiveformalismstodesigndistributeddataminingapplicationandsupporttheirefficient execution．Thispaperproposesanovelservice—orienteddaat mingingenginebasedoncloudcomputingrfamework， whichisnamedasCloudDM ．Differentiatingfrom grid—baseddistributeddataminingframework，CloudDM ex． ploitshtecapacityofopen sourcecloudcomputingplatfomr -- Hadoopofrlrage—scaledataanalysis，supportshte designandexecutionofdistributeddataminingapplicationsaccordingtoSOA (service．orientedarchitecture)． Moreover,itdiscussesandreportsthekeycomponentfunctionsandimplementationtechnologies．Accordingtohte designprinciplesofSOA anddaat mi ningenginebasedoncloudcomputing，htepaperCna solvehteproblemsin massivedataminingsystems，suchasbigdatastorage，daatprocessingandinteractiveoperationofalgorithms，etc．