{"title":"Congestion duration in hybrid ATM/TDMA networks","authors":"M. Talla, M. Kadoch","doi":"10.1109/GLOCOM.1998.776611","DOIUrl":null,"url":null,"abstract":"Congestion duration is analyzed for an ATM multiplexer node with a new global congestion control scheme. Congestion duration is defined as the number of consecutive frames lost due to congestion. The global congestion control scheme is based on the leaky bucket and virtual leaky bucket techniques. It prioritizes transit traffic over local traffic and ensures quality of service (QOS) to several classes of traffic (i.e., video, voice and data, etc.). Early warning cells with congestion criteria are transmitted periodically to all adjacent nodes. The transport users modulate end-to-end flow control parameters, i.e., protocol data unit size in case of video and voice users, and window size in case of data users, based on the instantaneous global congestion status. Simple analytical formula is derived for congestion duration. The leaky bucket scheme provided better congestion performance compared to virtual leaky bucket scheme.","PeriodicalId":414137,"journal":{"name":"IEEE GLOBECOM 1998 (Cat. NO. 98CH36250)","volume":"154 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"1998-11-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"IEEE GLOBECOM 1998 (Cat. NO. 98CH36250)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/GLOCOM.1998.776611","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0
Abstract
Congestion duration is analyzed for an ATM multiplexer node with a new global congestion control scheme. Congestion duration is defined as the number of consecutive frames lost due to congestion. The global congestion control scheme is based on the leaky bucket and virtual leaky bucket techniques. It prioritizes transit traffic over local traffic and ensures quality of service (QOS) to several classes of traffic (i.e., video, voice and data, etc.). Early warning cells with congestion criteria are transmitted periodically to all adjacent nodes. The transport users modulate end-to-end flow control parameters, i.e., protocol data unit size in case of video and voice users, and window size in case of data users, based on the instantaneous global congestion status. Simple analytical formula is derived for congestion duration. The leaky bucket scheme provided better congestion performance compared to virtual leaky bucket scheme.