Abstract

Most congestion control algorithms try to emulate processor sharing (PS) by giving each competing flow an equal share of a bottleneck link. This approach leads to fairness, and prevents long flows from hogging resources. For example, if a set of flows with the same round trip time share a bottleneck link, TCP’s congestion control mechanism tries to achieve PS; so do most of the proposed alternatives, such as eXplicit Control Protocol (XCP). But although they emulate PS well in a static scenario when all flows are long-lived, they do not come close to PS when new flows arrive randomly and have a finite amount of data to send, as is the case in today’s Internet. Typically, flows take an order of magnitude longer to complete with TCP or XCP than with PS, suggesting large room for improvement. And so in this paper, we explore how a new congestion control algorithm — Rate Control Protocol (RCP) — comes much closer to emulating PS over a broad range of operating conditions. In RCP, a router assigns a single rate to all flows that pass through it. The router does not keep flow-state, and does no per-packet calculations. Yet we are able to show that under a wide range of traffic characteristics and network conditions, RCP’s performance is very close to ideal processor sharing. Document type: Part of book or chapter of book

Abstract

State Estimation is an intrinsic element of many network manage-ment systems, like Power Distribution Networks and Water Distribution Networks, where its implementation not only facilitates real-time online monitoring with better observability, but it also enables an advanced control with improved system security. This paper presents a new technique based on State Estimation to address some general shortcomings of the current Active Queue Management schemes such as RED and discusses potential issues in TCP networks in order to achieve better performance. Document type: Part of book or chapter of book

Abstract

Several new protocols such as RBUDP, User-Level UDP, Tsunami, and SABUL, have been proposed as alternatives to TCP for high-speed data transfer. The purpose of this paper is to analyze experimentally the effects of SABUL congeston control algorithm on SABUL and performance metrics such as bandwidth utilization, self-fairness, and aggressiveness. Our results confirm some expected behavior of SABUL and reveal some less expected one. Our experiments also indicate that SABUL implementation and design can result in an even more erratic behavior and degraded performance under high-congestion conditions. Document type: Part of book or chapter of book