Using crowdsourcing, big data and machine (2013)

ID:18440542

大小:1.87 MB

页数:15页

时间:2018-09-18

Using crowdsourcing, big data and machine (2013)_第1页
Using crowdsourcing, big data and machine (2013)_第2页
Using crowdsourcing, big data and machine (2013)_第3页
Using crowdsourcing, big data and machine (2013)_第4页
Using crowdsourcing, big data and machine (2013)_第5页
资源描述:

《Using crowdsourcing, big data and machine (2013)》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、CollectiveMind:cleaninguptheresearchandexperimentationmessincomputerengineeringusingcrowdsourcing,bigdataandmachinelearningGrigoriFursinINRIA,FranceGrigori.Fursin@cTuning.orgAbstractSoftwareandhardwareco-designandoptimizationofHPCsystemshasbe-comeint

2、olerablycomplex,ad-hoc,timeconsuminganderrorproneduetoenor-mousnumberofavailabledesignandoptimizationchoices,complexinteractionsbetweenallsoftwareandhardwarecomponents,andmultiplestrictrequirementsplacedonperformance,powerconsumption,size,reliability

3、andcost.Wepresentournovellong-termholisticandpracticalsolutiontothisproblembasedoncustomizable,plugin-based,schema-free,heterogeneous,open-sourceCollectiveMindrepositoryandinfrastructurewithunifiedwebinterfacesandon-lineadvisesystem.Thiscollaborativef

4、rameworkdistributesanalysisandmulti-objectiveoff-lineandon-lineauto-tuningofcomputersystemsamongmanypar-ticipantswhileutilizinganyavailablesmartphone,tablet,laptop,clusterordatacenter,andcontinuouslyobserving,classifyingandmodelingtheirrealisticbehav

5、-ior.AnyunexpectedbehaviorisanalyzedusingshareddataminingandpredictivemodelingpluginsorexposedtothecommunityatcTuning.orgforcollaborativeexplanation,top-downcomplexityreduction,incrementalproblemdecompositionanddetectionofcorrelatingprogram,architect

6、ureorrun-timeproperties(features).hal-00850880,version1-10Aug2013Graduallyincreasingoptimizationknowledgehelpstocontinuouslyimproveop-timizationheuristicsofanycompiler,predictoptimizationsfornewprogramsorsuggestefficientrun-time(online)tuningandadapta

7、tionstrategiesdependingonend-userrequirements.Wedecidedtoshareallourpastresearchartifactsinclud-inghundredsofcodelets,numericalapplications,datasets,models,universalex-perimentalanalysisandauto-tuningpipelines,self-tuningmachinelearningbasedmetacompi

8、ler,andunifiedstatisticalanalysisandmachinelearningpluginsinapublicrepositorytoinitiatesystematic,reproducibleandcollaborativeresearch,developmentandexperimentationwithanewpublicationmodelwhereexperi-mentsandtechniquesarevalidated,rankedandimprovedbyt

当前文档最多预览五页,下载文档查看全文

此文档下载收益归作者所有

当前文档最多预览五页,下载文档查看全文
正文描述:

《Using crowdsourcing, big data and machine (2013)》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、CollectiveMind:cleaninguptheresearchandexperimentationmessincomputerengineeringusingcrowdsourcing,bigdataandmachinelearningGrigoriFursinINRIA,FranceGrigori.Fursin@cTuning.orgAbstractSoftwareandhardwareco-designandoptimizationofHPCsystemshasbe-comeint

2、olerablycomplex,ad-hoc,timeconsuminganderrorproneduetoenor-mousnumberofavailabledesignandoptimizationchoices,complexinteractionsbetweenallsoftwareandhardwarecomponents,andmultiplestrictrequirementsplacedonperformance,powerconsumption,size,reliability

3、andcost.Wepresentournovellong-termholisticandpracticalsolutiontothisproblembasedoncustomizable,plugin-based,schema-free,heterogeneous,open-sourceCollectiveMindrepositoryandinfrastructurewithunifiedwebinterfacesandon-lineadvisesystem.Thiscollaborativef

4、rameworkdistributesanalysisandmulti-objectiveoff-lineandon-lineauto-tuningofcomputersystemsamongmanypar-ticipantswhileutilizinganyavailablesmartphone,tablet,laptop,clusterordatacenter,andcontinuouslyobserving,classifyingandmodelingtheirrealisticbehav

5、-ior.AnyunexpectedbehaviorisanalyzedusingshareddataminingandpredictivemodelingpluginsorexposedtothecommunityatcTuning.orgforcollaborativeexplanation,top-downcomplexityreduction,incrementalproblemdecompositionanddetectionofcorrelatingprogram,architect

6、ureorrun-timeproperties(features).hal-00850880,version1-10Aug2013Graduallyincreasingoptimizationknowledgehelpstocontinuouslyimproveop-timizationheuristicsofanycompiler,predictoptimizationsfornewprogramsorsuggestefficientrun-time(online)tuningandadapta

7、tionstrategiesdependingonend-userrequirements.Wedecidedtoshareallourpastresearchartifactsinclud-inghundredsofcodelets,numericalapplications,datasets,models,universalex-perimentalanalysisandauto-tuningpipelines,self-tuningmachinelearningbasedmetacompi

8、ler,andunifiedstatisticalanalysisandmachinelearningpluginsinapublicrepositorytoinitiatesystematic,reproducibleandcollaborativeresearch,developmentandexperimentationwithanewpublicationmodelwhereexperi-mentsandtechniquesarevalidated,rankedandimprovedbyt

显示全部收起
温馨提示:
1. 部分包含数学公式或PPT动画的文件,查看预览时可能会显示错乱或异常,文件下载后无此问题,请放心下载。
2. 本文档由用户上传,版权归属用户,天天文库负责整理代发布。如果您对本文档版权有争议请及时联系客服。
3. 下载前请仔细阅读文档内容,确认文档内容符合您的需求后进行下载,若出现内容与标题不符可向本站投诉处理。
4. 下载文档时可能由于网络波动等原因无法下载或下载错误,付费完成后未能成功下载的用户请联系客服处理。
关闭