{"title":"Design of priority schemes in CSMA/CD local area networks","authors":"Yih-Long Chang, Sheldon Shen","doi":"10.1145/41824.41828","DOIUrl":null,"url":null,"abstract":"This paper compares the performance of three message based priority protocols in CSMA/CD local area networks. The major thesis is that traffic loads and message ratios are critical factors in network system performance. Thus, a message based priority protocol should not be designed based on any assumption on traffic and ratios. Since the dynamic protocol, proposed by Shen and Liu [12], is traffic and ratio independent, we argued that it should outperform the other two protocols, given a wide range of traffic and ratios. We perform 600 experiments and the statistics indeed confirms our contention, with a chance of errors in most cases below only .01%.","PeriodicalId":186490,"journal":{"name":"Annual Simulation Symposium","volume":"1 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"1987-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"8","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Annual Simulation Symposium","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1145/41824.41828","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 8
Abstract
This paper compares the performance of three message based priority protocols in CSMA/CD local area networks. The major thesis is that traffic loads and message ratios are critical factors in network system performance. Thus, a message based priority protocol should not be designed based on any assumption on traffic and ratios. Since the dynamic protocol, proposed by Shen and Liu [12], is traffic and ratio independent, we argued that it should outperform the other two protocols, given a wide range of traffic and ratios. We perform 600 experiments and the statistics indeed confirms our contention, with a chance of errors in most cases below only .01%.