{"title":"An investigation into performance of congestion control mechanisms in ATM-UBR service for TCP sources","authors":"J. Phunitatkul, T. Erke","doi":"10.1109/ICON.2001.962386","DOIUrl":null,"url":null,"abstract":"The performance of congestion control mechanisms in UBR service for TCP sources is investigated. Not only static TCP sources but also dynamic TCP sources are applied to investigate the performance of congestion control in the practical way. For static sources, the result from random early detection (RED) achieves high efficiency but not the highest fairness. RED improves fairness slightly less than FBA. For dynamic sources, the efficiency is not much different for each UBR drop policy according to the different traffic profiles. RED performs the highest fairness. The difference of implementation between the other schemes and RED as the other schemes are most appropriate to implement with the instantaneous queue length, but that RED is appropriate to implement with the average queue length which is suitable for the real or dynamic Internet traffic characteristic.","PeriodicalId":178842,"journal":{"name":"Proceedings. Ninth IEEE International Conference on Networks, ICON 2001.","volume":"38 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2001-10-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"3","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Proceedings. Ninth IEEE International Conference on Networks, ICON 2001.","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/ICON.2001.962386","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 3
Abstract
The performance of congestion control mechanisms in UBR service for TCP sources is investigated. Not only static TCP sources but also dynamic TCP sources are applied to investigate the performance of congestion control in the practical way. For static sources, the result from random early detection (RED) achieves high efficiency but not the highest fairness. RED improves fairness slightly less than FBA. For dynamic sources, the efficiency is not much different for each UBR drop policy according to the different traffic profiles. RED performs the highest fairness. The difference of implementation between the other schemes and RED as the other schemes are most appropriate to implement with the instantaneous queue length, but that RED is appropriate to implement with the average queue length which is suitable for the real or dynamic Internet traffic characteristic.