!!!!!A Performance Comparison of Monofractal and Multifractal Traffic Streams (2002)

!!!!!A Performance Comparison of Monofractal and Multifractal Traffic Streams (2002)

ID:39994427

大小:96.06 KB

页数:10页

时间:2019-07-16

!!!!!A Performance Comparison of Monofractal and Multifractal Traffic Streams (2002)_第1页
!!!!!A Performance Comparison of Monofractal and Multifractal Traffic Streams (2002)_第2页
!!!!!A Performance Comparison of Monofractal and Multifractal Traffic Streams (2002)_第3页
!!!!!A Performance Comparison of Monofractal and Multifractal Traffic Streams (2002)_第4页
!!!!!A Performance Comparison of Monofractal and Multifractal Traffic Streams (2002)_第5页
资源描述:

《!!!!!A Performance Comparison of Monofractal and Multifractal Traffic Streams (2002)》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、APerformanceComparisonofMonofractalandMultifractalTrafficStreamsRamBalakrishnanCareyWilliamsonDepartmentofComputerScienceUniversityofSaskatchewanE-mail:carey@cs.usask.caAbstractOneapproachformodelingmonofractal(i.e.,self-similar)trafficisproposedbyNorros[1

2、0].ThisapproachThispaperfocusesonthedifferencesbetweenmonofrac-usesthreeparameters–themeanbitrate(inunitsoftalandmultifractalnetworktraffic,bothintermsofmod-bits/sec),theNorrosvariancecoefficient(thevariance-to-elingapproaches,andintermsofpracticalimpact

3、sonnet-meanratioofthetrafficatthe1.0-secondtimescale,inunitsworkperformance.Empiricaltraffictracesareusedintheofbit-sec),andtheHurstparameter(aunit-lessvaluebe-parameterizationandevaluationoftrafficmodels.Simu-tween0.5and1.0,whichquantifiestheglobalscalingp

4、rop-lationisusedtoevaluatetheperformancedifferencesbe-ertiesofthetrafficacrosstimescales,andhencethedegreetweenmonofractalandmultifractaltraffic,bothforindivid-oflong-rangedependence)–forcharacterizingself-similarualtrafficstreamsandforaggregationsoftraffics

5、treams.datatraffic[10,17].However,onedrawbackoftheNor-Thesimulationresultsindicatethatmultifractaltrafficof-rostrafficmodelisthatitdoesnotcharacterizeshort-rangefersgreaterpotentialformultiplexinggains,bothwithincorrelations[1],whichhavebeenshowninthelitera

6、turetoandacrosssources.Thesensitivityoftheseresultstosourcehavesignificantimpactonthequeueingbehavior(andthuscharacteristicsandswitchbuffersizeisalsoexplored.thebufferoverflowprocess)inpracticalnetworks[6,7,13].MorerecentnetworktrafficanalyseshaveshownthatW

7、AN(WideAreaNetwork)TCP/IP(TransmissionControl1.IntroductionProtocol/InternetProtocol)trafficismultifractalinnature.Thatis,thetrafficexhibitsdifferentscalingcharacteristicsatlargetimescales(e.g.,hundredsofmillisecondsormore)Accuratetrafficmodelingiscrucialfo

8、rnetworkperfor-thanatsmallertimescales[3,5].mancestudies.Highspeedmultimedianetworksexhibitcomplextopologies,heterogeneoustransmissionspeeds,TocapturethemultifractalcharacteristicsofnetworkandawiderangeofQoSrequirements.Co

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。