J. Khamse-Ashari, G. Kesidis, I. Lambadaris, B. Urgaonkar, Yiqiang Q. Zhao
{"title":"Max-min Fair scheduling of variable-length packet-flows to multiple servers by deficit round-robin","authors":"J. Khamse-Ashari, G. Kesidis, I. Lambadaris, B. Urgaonkar, Yiqiang Q. Zhao","doi":"10.1109/CISS.2016.7460534","DOIUrl":null,"url":null,"abstract":"We describe a scheduler based on deficit-round robin (DRR) for multiple servers of multiple packet-flows, where each packet-flow may be served by only a subset of available (preferred) servers. The scheduler uses a token allocation algorithm that is weighted max-min fair, and so we've called it Multi-Server Max-min Fair DRR (MSMF-DRR). The scheduler also compensates for potential errors in estimates of server capacities when determining token allocations, and considers service underflow resulting in unused tokens at the end of a round. Numerical examples are given to illustrate how the scheduler itself is weighted max-min fair.","PeriodicalId":346776,"journal":{"name":"2016 Annual Conference on Information Science and Systems (CISS)","volume":"82 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2016-03-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"9","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2016 Annual Conference on Information Science and Systems (CISS)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/CISS.2016.7460534","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 9
Abstract
We describe a scheduler based on deficit-round robin (DRR) for multiple servers of multiple packet-flows, where each packet-flow may be served by only a subset of available (preferred) servers. The scheduler uses a token allocation algorithm that is weighted max-min fair, and so we've called it Multi-Server Max-min Fair DRR (MSMF-DRR). The scheduler also compensates for potential errors in estimates of server capacities when determining token allocations, and considers service underflow resulting in unused tokens at the end of a round. Numerical examples are given to illustrate how the scheduler itself is weighted max-min fair.