简介：(DVR)直接体积显示是允许用户有效地探索并且学习容量的数据集的一种强大的可视化技术。不同透明性设置能灵活地被分到不同结构以便一些珍贵信息能在直接体积被揭示显示的图象(DVRI)。然而，因为他们不知道任何重要信息是否从DVRI的透明区域是失踪的，经常，一些风险总是是的感觉与DVR联系了的最终用户。在这份报纸，我们调查怎么半自动地产生一套DVRI并且也，能揭示信息的一张动画在原来的DVRI并且同时错过了满足象连贯那样的一些图象质量标准。一个完全的框架被开发处理与产生和可见性知道的DVRI和动画的优秀评估有关的各种各样的问题。我们的技术能减少直接使用体积显示的风险并且因此在显示系统的体积增加用户的信心。

简介：TheMonte-CarloArrayProcessor(MAP)hasbeendesignedusingcommodityofftheshelf(COTS)itemstoprovidetheCPUrequirementsoffulleventsimulationfortheLHCexperiments.Thesolutionishowevercompletelygeneral,soandCPUintensiveapplicationwithlimitedinputrequirementscanberunonthesystem.Operatingcontrolsoftwarehasbeenwrittentomanagethedataflowoiverthe100BaseTethernetconnectingthe300nodes(400MHzPII's)tothe6mastercontrolnodes700MHzPIII'seachwith500Gbofdisk),Upgradeto1000nodesisplkanned.Jobcontrolsoftwarethatallowstheusertorunthesamejobonallnodes,whilstallowingforsmalldifferencesininitialisationparametersbetweennodeshasalsobeenwritten.GMAPistheGRIDawareMAPcontrolsoftware,Thisallowsremotejobpreparationandsubmissionusingglobustoolkitforauthentificationandcommunication.ThesoftwarewillbeavailableandopensthepossibilityfordoingmassiveMonteCarloproductionoverseveralremoteMAPsitessimultaneously.

简介：ThispapergivesanoverallintroductiontothebasicconceptofLAC(location-awarecomputing)anditsdevelopmentstatus,putsforwardanintegratedlocation-awarecomputingarchitecturewhichisusefulfordesigningthereasonablelogicalmodelofLBS(location-basedservice).Finally,abriefintroductionisconductedonaLACexperimentalprototype,whichactsasamobileurbantourismassistant.

简介：WirelessMeshNetworks(WMNs)areenvisionedtosupportthewiredbackbonewithawirelessBackboneNetworks(BNet)forprovidinginternetconnectivitytolarge-scaleareas.Withawiderangeofinternet-orientedapplicationswithdifferentQualityofService(QoS)requirement,thelarge-scaleWMNsshouldhavegoodscalabilityandlargebandwidth.Inthispaper,aLoadAwareAdaptiveBackboneSynthesis(LAABS)algorithmisproposedtoautomaticallybalancethetrafficflowintheWMNs.TheBNetwilldynamicallysplitintosmallersizeormergeintobiggeroneaccordingtostatisticloadinformationofBackboneNodes(BNs).SimulationresultsshowLAABSgeneratesmoderateBNetsizeandconvergesquickly,thusprovidingscalableandstableBNettofacilitatetrafficflow.

简介：在最近的年里，与数据的快速的开发集中的应用程序，数据复制成为了一种创新技术让数据格子改进数据可获得性，和还原剂文件转移时间和带宽消费。复制品的放置被证明了是必须被解决认识到数据复制的过程的最困难的问题。这份报纸探讨服务(QoS)的质量在数据格子的知道的复制品放置问题，并且建议动态编程基于不仅让一个QoS要求保证全面复制花费了，而且能最小化的复制品放置算法，包括存储费用和通讯费用。由模拟，实验证明复制品放置算法超过存在在数据的流行复制品放置技术格子。

简介：Anopticalsignal-to-noiseratio(OSNR)awarelightpathprovisioningmechanism(OSNR-LPM)isproposedfordistributedopticalnetworks.ThisOSNR-LPMtakestheOSNRvalueofthelightwavealongthelightpathintoconsiderationwhenestablishingthelightpathfortheconnectionrequestusingresourcereservationprotocol-trafficengineering(RSVP-TE).Moreover,theOSNR-LPMmakesfulladvantagesoftheOSNRmonitoringfunctionineachnodeandassignsthelightwavebyjudgingtheOSNRvaluecarriedbythesignalingmessageinordertoguaranteethereliableestablishmentofthelightpathinthedataplane.ThesimulationresultsshowthattheOSNR-LPMoutperformsotherlightpathassignmentmechanismsintermsofrealblockingprobabilityinthedataplane.

简介：Thenetworks-on-chip(NoC)communicationhasanincreasinglylargerimpactonthesystempowerconsumptionandperformance.Emergingtechnologies,likesurfacewave,arebelievedtohavelowertransmissionlatencyandpowerconsumptionovertheconventionalwirelessNoC.Therefore,thispaperstudieshowtooptimizethenetworkperformanceandpowerconsumptionbygivingthepacket-switchingfabricandtrafficpatternofeachapplication.Comparedwiththeconventionalmethodofwire-linked,whichaddswirelesstransceiversbyusingthegeneticalgorithm(GA),theproposedmaximaldecliningsortingalgorithm(MDSA)caneffectivelyreducetimeconsumptionbyasmuchas20.4%to35.6%.Wealsoevaluatethepowerconsumptionandconfigurationtimetoprovetheeffectiveoftheproposedalgorithm.

简介：在这份报纸，我们认为传播为multicell策划的干扰知道的uplink多重输入多重产量(MIMO)系统。不同于没有考虑可能引起到另外的房间的干扰的常规传播计划，我们联合优化收发器beamforming向量当移开intercell干扰时，最大化需要的信号。为每个传输器与二条天线被装备的一个二房间的系统，明确地，我们为经由generalized-eigen分析称为协调beamforming(CBF)的transmit计划导出靠近形式的表达式。而且当不对称的干扰被考虑时，我们给平衡beamforming(BBF)，策划介入的传输器在哪儿打击在最大化需要的信号并且最小化产生干扰之间的妥协。模拟结果证明两个计划在不同情形下面比常规计划更好表现。

简介：ToimproveuserexperienceofcompositeWebservices,auser-awarequalityofservice(QoS)basedWebservicescompositionmodelisproposed.Undersuchmodel,aWebservicesselectionmethodbasedonquantumgeneticalgorithmisproposed.Thisalgorithmusesquantumbitencoding,dynamicstep-lengthquantumgateangleadjustment,neighborhoodservicesearchanddynamicpunishmentstrategytoexpandsearchscopeandspeedupconvergence.SimulationexperimentshowsthatthisalgorithmismoreefficientthanotherexistingalgorithmsinWebservicesselection.

简介：Curativetherapyforspinalcordinjury(SCI)remainselusive,howeveridentifyingoptionsfortailoredtreatmentstrategiesisinfullswing.Likeinthebraintherearedistinctregionsintheadultspinalcordthatharborneuralprogenitorcells(NPCs)(Horneretal.,2000).Thisoffersthepossibilityofrecruitingthesecellsinreparativeapproachestosupportendogenousspinalcordregenerativecapacities.Hereby,one

简介：

简介：Timeefficiencyofkeyestablishmentandupdateisoneofthemajorproblemscontributorykeymanagementsstrivetoaddress.Toachievebettertimeefficiencyinkeyestablishment,weproposeaLocation-basedHuffman(L-Huffman)scheme.First,usersareseparatedintoseveralsmallgroupstominimizecommunicationcostwhentheyaredistributedoverlargenetworks.Second,bothuser'scomputationdifferenceandmessagetransmissiondelayaretakenintoconsiderationwhenHuffmancodingisemployedtoformingtheoptimalkeytree.Third,thecombinedweightsinHuffmantreearelocatedinahigherplaceofthekeytreetoreducethevarianceoftheaveragekeygenerationtimeandminimizethelongestkeygenerationtime.SimulationsdemonstratethatL-HuffmanhasmuchbetterperformanceinwideareanetworksandisalittlebetterinlocalareanetworkthanHuffmanscheme.

简介：图象compositing广泛地被用来从分开的来源图象把视觉元素合为一幅单个图象。尽管最近的图象compositing技术能够完成从不同来源的视觉元素的光滑的相配，他们中的大多数含蓄地假设来源图象在一样的观点被拿。在这份报纸，我们在场从多重来源的compositing小说图象目标的一条途径想象哪个有不同观点。我们的关键想法是为来源图象目标的有意义的部件构造3D代理，并且使用这些3D部件代理在一样的观点一起弄歪并且无缝地合并部件。认识到这个想法，我们介绍一个并列框架的基于的单个看法的照相机刻度算法处理图象对象的一般类型，得到长方体的一个结构知道的长方体优化算法正确结构关系地为图象对象部件代理，并且最后，3D代理转变指导了图象变弯缝对象部件的算法。我们进一步基于这条compositing途径描述一个新奇应用程序自动地综合从一套模范的很多图象对象。试验性的结果证明我们的compositing途径能被用于许多图象目标，例如椅子，杯，灯，和机器人，并且合成申请能从模范的一个小集合与重要形状和风格变化创造新奇图象目标。

简介：Quality-of-Service(QoS)描述万维网服务的非功能的特征。因此，QoS是在服务选择，作文和容错的一个批评参数，并且必须被QoS预言方法的某类型精确地决定。随万维网服务的数字的戏剧的增加，然而，数据稀疏引起的预言失败成为了批评挑战。新‘hybriduser-location-aware预言基于加权的Adamic-Adar(WAA)’(HUWAA)被建议。含蓄的邻居搜索被合并地点因素优化。同时，改进算法的能力在公共真实世界数据集的实验被验证解决数据稀少问题。新算法超过基于条款的pearson关联系数(IPCC)存在，基于用户的pearson关联系数(UPCC)和万维网服务recommender系统(WSRec)算法。

简介：InvehicularAd-hocnetwork(VANET),manymulti-hopbroadcastschemesareemployedtowidelypropagatethewarningmessagesamongvehiclesandthekeyistodynamicallydeterminetheoptimalrelayvehicleforretransmission.Inordertoachievereliableandfastdeliveryofwarningmessages,thispaperproposesadelay-awareandreliablebroadcastprotocol(DR-BP)basedontransmitpowercontroltechnique.First,acomprehensivemodelisderivedtoevaluatethetransmissioninvehicle-to-vehiclecommunications.Thismodelconsidersthewirelesschannelfading,transmissiondelayandretransmissionscharactersoccurringinthephysicallayer/mediumaccesscontrol(PHY/MAC)layer.Then,alocaloptimalrelayselectionmechanismbasedontheabovemodelisdesigned.InDR-BPscheme,onlythevehicleselectedastheoptimalrelayscanforwardwarningmessagesandthetransmitpoweristime-varying.Finally,extensivesimulationsverifytheperformanceofDR-BPunderdifferenttrafficscenarios.SimulationresultsshowthatDR-BPoutperformsthetraditionalslotted1-persistence(S1P)andfloodingschemeintermsofpacketsdeliveryratioandtransmissiondelay.

简介：Theabsenceofnetworkinfrastructureandopportunisticspectrumaccessincognitiveradioadhocnetworks(CRAHNs)resultsinconnectivityandstabilityproblems.Clusteringisknownasaneffectivetechniquetoovercomethisproblem.Clusteringimprovesnetworkperformancebyimplementingalogicalnetworkbackbone.Therefore,howtoefficientlyconstructthisbackboneamongCRAHNsisofinterest.Inthispaper,weproposeanewclusteringalgorithmforCRAHNs.Moreover,wemodelanovelclusterheadselectionfunctionbasedonthechannelheterogeneityintermoftransmissionranges.Tothebestofourknowledge,thisisthefirstattempttomodelthechannelheterogeneityintotheclusteringformationincognitiveradionetworks.Simulationresultsshowthattheperformanceofclusteringissignificantlyimprovedbythechannelheterogeneityconsiderations.

简介：Thewidespreadoflocation-basedsocialnetworksbringsaboutahugevolumeofusercheck-indata,whichfacilitatestherecommendationofpointsofinterest（POIs）.Recentadvancesondistributedrepresentationshedlightonlearninglowdimensionaldensevectorstoalleviatethedatasparsityproblem.CurrentstudiesonrepresentationlearningforPOIrecommendationembedbothusersandPOIsinacommonlatentspace,andusers＇preferenceisinferredbasedonthedistance/similaritybetweenauserandaPOI.SuchanapproachisnotinaccordancewiththesemanticsofusersandPOIsastheyareinherentlydifferentobjects.Inthispaper,wepresentanoveltranslation-based,timeandlocationaware（TransTL）representation,whichmodelsthespatialandtemporalinformationasarelationshipconnectingusersandPOIs.Ourmodelgeneralizestherecentadvancesinknowledgegraphembedding.Thebasicideaisthattheembeddingofa〈time,location〉paircorrespondstoatranslationfromembeddingsofuserstoPOIs.SincethePOIembeddingshouldbeclosetotheuserembeddingplustherelationshipvector,therecommendationcanbeperformedbyselectingthetop-kPOIssimilartothetranslatedPOI,whichareallofthesametypeofobjects.Weconductextensiveexperimentsontworeal-worlddata.sets.TheresultsdemonstratethatourTransTLmodelachievesthestate-of-the-artperformance.Itisalsomuchmorerobusttodatasparsitythanthebaselines.

简介：

简介：Themulticoreevolutionhasstimulatedrenewedinterestsinscalingupapplicationsonshared-memorymultiprocessors,significantlyimprovingthescalabilityofmanyapplications.Butthescalabilityislimitedwithinasinglenode;thereforeprogrammersstillhavetoredesignapplicationstoscaleoutovermultiplenodes.Thispaperrevisitsthedesignandimplementationofdistributedsharedmemory(DSM)asawaytoscaleoutapplicationsoptimizedfornon-uniformmemoryaccess(NUMA)architectureoverawell-connectedcluster.ThispaperpresentsMAGI,anefficientDSMsystemthatprovidesatransparentsharedaddressspacewithscalableperformanceonaclusterwithfastnetworkinterfaces.MAGIisuniqueinthatitpresentsaNUMAabstractiontofullyharnessthemulticoreresourcesineachnodethroughhierarchicalsynchronizationandmemorymanagement.MAGIalsoexploitsthememoryaccesspatternsofbig-dataapplicationsandleveragesasetofoptimizationsforremotedirectmemoryaccess(RDMA)toreducethenumberofpagefaultsandthecostofthecoherenceprotocol.MAGIhasbeenimplementedasauser-spacelibrarywithpthread-compatibleinterfacesandcanrunexistingmultithreadedapplicationswithminimizedmodifications.WedeployedMAGIoveran8-nodeRDMA-enabledcluster.ExperimentalevaluationshowsthatMAGIachievesupto9.25:4speedupcomparedwithanunoptimizedimplementation,leadingtoasealableperformanceforlarge-scaledata-intensiveapplications.

简介：how to balance the masculinity/femininity from different culture and background in order to maximize the team power is worth considering by managers.The above four dimensions illuminated the most important cultural differences that affect on business. International managers should be able to aware not only the cultural difference but also the intercultural communication.Gudykunst and Kim (1992摘要,the power distance can be viewed as the subordinates respect the superiors.That is the corporate culture in orient. Let us take a look at the western way. It is not a whole converse phenomenon. There is 'the Eiffel Tower culture' (Trompernaars,the international managers must pay attention to the clashes and be aware of. How to work the subordinates together efficiently and more cooperatively is important too.And then there is also a large discrepancy on the uncertainty avoidance. (Hoecklin