architecture for stateful data-intensive analytics

《architecture for stateful data-intensive analytics》由会员上传分享，免费在线阅读，更多相关内容在学术论文-天天文库。

1、UNIVERSITYOFCALIFORNIA,SANDIEGOArchitecturesforStatefulData-intensiveAnalyticsAdissertationsubmittedinpartialsatisfactionoftherequirementsforthedegreeDoctorofPhilosophyinComputerSciencebyDionysiosLogothetisCommitteeincharge:KennethYocum,ChairReneCruzAlinDeutschMassimoFranceschettiAl

2、exSnoerenGeoffreyM.Voelker2011CopyrightDionysiosLogothetis,2011Allrightsreserved.ThedissertationofDionysiosLogothetisisapproved,anditisacceptableinqualityandformforpublicationonmicrofilmandelectronically:ChairUniversityofCalifornia,SanDiego2011iiiTABLEOFCONTENTSSignaturePage..........

3、........................iiiTableofContents.................................ivListofFigures..................................viiListofTables...................................xiAcknowledgements................................xiiVita........................................xiiiAbstract

4、oftheDissertation...........................xivChapter1Introduction............................11.1The“bigdata”promise...................11.2Data-intensivecomputing..................31.3Statefuldata-intensivecomputing.............51.3.1Anexample:webcrawlqueue...........61.3.2Challen

5、gesinmanagingstate...........71.4Architecturesforstatefulanalytics.............81.4.1ContinuousETLanalytics.............91.4.2Statefulbulkdataprocessing............101.5Contributions........................111.6Outline............................12Chapter2Backgroundandrelatedwork...

6、...............142.1Groupwiseprocessing....................142.2TheMapReduceprogrammingmodel...........152.3MapReducearchitecture..................182.3.1Programexecution.................182.3.2Faulttolerance...................202.4Statefulanalyticsonunstructureddata..........212.4.

7、1Automaticincrementalization...........212.4.2Iterativeanalytics..................222.4.3Modelsforincrementalanalytics..........23Chapter3Statefulonlineanalytics.....................253.1Design............................273.1.1ContinuousMapReduce..............293.1.2Programexecutio

8、n.................2