{"title":"Assessment and performance evaluation of PCE-based inter-layer traffic engineering","authors":"S. Gunreben, F. Rambach","doi":"10.1109/ONDM.2008.4578399","DOIUrl":null,"url":null,"abstract":"Multi-layer (ML) and multi-domain networks require path computation elements (PCE) for constraint-based path calculation. In this paper we introduce and evaluate qualitatively as well as quantitatively the PCE scenarios newly proposed by the IETF for PCE-based inter-layer traffic engineering. Requirements on additional communication, on hardware and on optimality of path computation serve as the qualitative metrics in our comparison. The path setup delay is derived analytically and serves as the quantitative metric. We derive the results using simulations on the ML German reference network with 17 nodes and back-up our results by two different ML TE routing algorithms. We show that one single ML PCE performs best in the overall qualitatively and quantitatively comparison.","PeriodicalId":155835,"journal":{"name":"2008 International Conference on Optical Network Design and Modeling","volume":"7 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2008-03-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"15","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2008 International Conference on Optical Network Design and Modeling","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/ONDM.2008.4578399","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 15
Abstract
Multi-layer (ML) and multi-domain networks require path computation elements (PCE) for constraint-based path calculation. In this paper we introduce and evaluate qualitatively as well as quantitatively the PCE scenarios newly proposed by the IETF for PCE-based inter-layer traffic engineering. Requirements on additional communication, on hardware and on optimality of path computation serve as the qualitative metrics in our comparison. The path setup delay is derived analytically and serves as the quantitative metric. We derive the results using simulations on the ML German reference network with 17 nodes and back-up our results by two different ML TE routing algorithms. We show that one single ML PCE performs best in the overall qualitatively and quantitatively comparison.