optimizing path query performance graph clustering strategies

optimizing path query performance graph clustering strategies

ID:34455518

大小:337.89 KB

页数:35页

时间:2019-03-06

optimizing path query performance graph clustering strategies_第1页
optimizing path query performance graph clustering strategies_第2页
optimizing path query performance graph clustering strategies_第3页
optimizing path query performance graph clustering strategies_第4页
optimizing path query performance graph clustering strategies_第5页
资源描述:

《optimizing path query performance graph clustering strategies》由会员上传分享,免费在线阅读,更多相关内容在教育资源-天天文库

1、OptimizingPathQueryPerformance:GraphClusteringStrategiesyzxYun-WuHuangNingJingElkeA.RundensteinerIBMT.J.WatsonLabsChangshaInstituteofTechnologyWorcesterPolytechnicInstituteywh@us.ibm.comjning@eecs.umich.edurundenst@cs.wpi.eduAbstractPathqueriesovertransportationnetworksareoperations

2、requiredbymanyGeographicInformationSystemsapplications.Suchnetworks,typicallymodeledasgraphscomposedofnodesandlinksandrepresentedaslinkrelations,canbeverylargeandhenceoftenneedtobestoredonsecondarystor-agedevices.PathquerycomputationoversuchlargepersistentnetworksamountstohighI/Ocost

3、sduetohavingtorepeatedlybringinlinksfromthelinkrelationfromsecondarystorageintothemainmemorybu erforprocessing.Thispaperisthe rsttopresentacomparativeexperimentalevaluationofalternativegraphclusteringsolutionsinordertoshowtheire ectivenessinpathqueryprocessingovertransportationnetwor

4、ks.Clusteringoptimizationisattractivebecauseitdoesnotincuranyrun-timecost,requiresnoauxiliarydatastructures,andiscomplimentarytomanyoftheexistingsolutionsonpathqueryprocessing.Inthispaper,wedevelopanovelclusteringtech-nique,calledspatialpartitionclustering(SPC),thatexploitsuniqueprop

5、ertiesoftransportationnetworkssuchasspatialcoordinatesandhighlocality.Weidentifyotherpromisingcandidatesforclusteringoptimizationsfromtheliterature,suchastwo-waypartitioningandapproximatetopo-logicalclustering.We ne-tunethemtooptimizetheirI/Obehaviorforpathqueryprocessing.Ourexperime

6、ntalevaluationoftheperformanceofthesegraphclusteringtechniquesusinganactualcityroadnetworkaswellasrandomlygeneratedgraphsconsidersvariationsinparameterssuchasmemorybu ersize,lengthofthepaths,locality,andout-degree.Ourexperimentalresultsarethefoundationforestablishingguidelinestoselec

7、tthebestclusteringtechniquebasedonthetypeofnetworks.We ndthatourSPCperformsthebestforthehighlyinterconnectedcitymap;thehybridapproachforrandomgraphswithhighlocality;andthetwo-waypartitioningbasedonlinkweightsforrandomgraphswithnolocality.IndexTerms

8、PathQueryProcessing,TransportationN

9、etworks,Spat

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
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。