{"title":"Performance of TCP with a Proportional Integral Enhanced (PIE) queue management policy","authors":"Santosh Chavan, G. Raina","doi":"10.1109/CCDC.2015.7162066","DOIUrl":null,"url":null,"abstract":"Bufferbloat refers to the problem of excessive queuing delays that are caused due to large, and unmanaged, router buffers in the Internet. The Proportional Integral Enhanced (PIE) queue policy has been proposed to explicitly control queuing delays in routers. In this paper, we study a fluid model for Compound TCP coupled with the PIE queue policy. A control-theoretic analysis of the fluid model highlights that it would be difficult to stabilize the system as the propagation delay, or the target queuing delay, gets large. We then computationally show that, as model parameters vary, loss of local stability may occur via a Hopf bifurcation. Finally, we conduct packet-level simulations to confirm the emergence of limit cycles, in the queue size, as the propagation delay or the target queuing delay increases.","PeriodicalId":273292,"journal":{"name":"The 27th Chinese Control and Decision Conference (2015 CCDC)","volume":"25 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2015-05-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"6","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"The 27th Chinese Control and Decision Conference (2015 CCDC)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/CCDC.2015.7162066","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 6
Abstract
Bufferbloat refers to the problem of excessive queuing delays that are caused due to large, and unmanaged, router buffers in the Internet. The Proportional Integral Enhanced (PIE) queue policy has been proposed to explicitly control queuing delays in routers. In this paper, we study a fluid model for Compound TCP coupled with the PIE queue policy. A control-theoretic analysis of the fluid model highlights that it would be difficult to stabilize the system as the propagation delay, or the target queuing delay, gets large. We then computationally show that, as model parameters vary, loss of local stability may occur via a Hopf bifurcation. Finally, we conduct packet-level simulations to confirm the emergence of limit cycles, in the queue size, as the propagation delay or the target queuing delay increases.