Results 1  10
of
15
A ControlTheoretic Approach to Flow Control
, 1991
"... This paper presents a controltheoretic approach to reactive flow control in networks that do not reserve bandwidth. We assume a roundrobinlike queue service discipline in the output queues of the network’s switches, and propose deterministic and stochastic models for a single conversation in a ne ..."
Abstract

Cited by 382 (7 self)
 Add to MetaCart
This paper presents a controltheoretic approach to reactive flow control in networks that do not reserve bandwidth. We assume a roundrobinlike queue service discipline in the output queues of the network’s switches, and propose deterministic and stochastic models for a single conversation in a network of such switches. These models motivate the PacketPair rate probing technique, and a provably stable ratebased flow control scheme. A Kalman state estimator is derived from discretetime state space analysis, but there are difficulties in using the estimator in practice. These difficulties are overcome by a novel estimation scheme based on fuzzy logic. We then present a technique to extract and use additional information horn the system to develop a continuoustime system model. This is used to design a wuisnt of the control law that is also provably stable, and, in addition, takes control action as rapidly as possible. Finally, practical issues such as correcting parameter drift and cmmlination with window flow control are described.
Connections with Multiple Congested Gateways in PacketSwitched Networks Part 1: Oneway Traffic
 ACM Computer Communication Review
, 1991
"... In this paper we explore the bias in TCP/IP networks against connections with multiple congested gateways. We consider the interaction between the bias against connections with multiple congested gateways, the bias of the TCP window modification algorithm against connections with longer roundtrip ti ..."
Abstract

Cited by 262 (14 self)
 Add to MetaCart
In this paper we explore the bias in TCP/IP networks against connections with multiple congested gateways. We consider the interaction between the bias against connections with multiple congested gateways, the bias of the TCP window modification algorithm against connections with longer roundtrip times, and the bias of Drop Tail and Random Drop gateways against bursty traffic. Using simulations and a heuristic analysis, we show that in a network with the window modification algorithm in 4.3 tahoe BSD TCP and with Random Drop or Drop Tail gateways, a longer connection with multiple congested gateways can receive unacceptably low throughput. We show that in a network with no bias against connections with longer roundtrip times and with no bias against bursty traffic, a connection with multiple congested gateways can receive an acceptable level of throughput. We discuss the application of several current measures of fairness to networks with multiple congested gateways, and show that diff...
RoundRobin Scheduling for MaxMin Fairness in Data Networks
 IEEE JOURNAL ON SELECTED AREAS IN COMMUNICATIONS
, 1991
"... This paper studies a simple strategy, proposed independently by Gallager [1] and Katevenis [2], for fairly allocating link capacity in a pointtopoint packet network with virtual circuit routing. Each link offers its packet transmission slots to its user sessions by polling them in roundrobin orde ..."
Abstract

Cited by 110 (0 self)
 Add to MetaCart
This paper studies a simple strategy, proposed independently by Gallager [1] and Katevenis [2], for fairly allocating link capacity in a pointtopoint packet network with virtual circuit routing. Each link offers its packet transmission slots to its user sessions by polling them in roundrobin order. In addition, window flow control is used to prevent excessive packet queues at the network nodes. As the window size increases, the session throughput rates are shown to approach limits that are perfectly fair in the maxmin sense. That is, the smallest session rate in the network is as large as possible and, subject to that constraint, the secondsmallest session rate is as large as possible, etc. If each session has periodic input (perhaps with jitter) or has such heavy demand that packets are always waiting to enter the network, then a finite window size suffices to produce perfectly fair throughput rates. The roundrobin method is considerably simpler than earlier strategies for achieving global fairness. The fair session rates are not explicitly computed, and the only overhead communication is that required for the window acknowledgments. The main drawback is that large windows are needed to achieve even approximately fair throughputs in some (hopefully rare) situations, and large windows permit large crossnetwork delays. Fortunately, the roundrobin method offers other throughput guarantees that, while falling short of perfect fairness, do apply even for sessions with small windows. Such sessions are promised reasonable bounds on their crossnetwork packet delay as well.
PacketPair Flow Control
 IEEE/ACM Transactions on Networking
, 1994
"... This paper presents the packetpair ratebased feedback flow control scheme. This scheme is designed for networks where individual connections do not reserve bandwidth and for the available bitrate (besteffort) component of integrated networks. We assume a roundrobinlike queue service discipline ..."
Abstract

Cited by 44 (2 self)
 Add to MetaCart
This paper presents the packetpair ratebased feedback flow control scheme. This scheme is designed for networks where individual connections do not reserve bandwidth and for the available bitrate (besteffort) component of integrated networks. We assume a roundrobinlike queue service discipline in the output queues of the network's switches, and propose a linear stochastic model for a single conversation in a network of such switches. These model motivates the PacketPair rate probing technique, which forms the basis for provably stable discrete and continuous time ratebased flow control schemes. We present a novel state estimation scheme based on fuzzy logic. We then address several practical concerns: dealing with system startup, retransmission and timeout strategy, and dynamic setpoint probing. We present a finite state machine as well as source code for a model implementation. The dynamics of a single source, the interactions of multiple sources, and the behavior of packetpai...
24) "Metrics for the Evaluation of Congestion Control Mechanisms
 Congestion Control in the RFC Series", M
, 2008
"... This memo provides information for the Internet community. It does not specify an Internet standard of any kind. Distribution of this memo is unlimited. IESG Note This document is not an IETF Internet Standard. It represents the individual opinion(s) of one or more members of the TMRG Research ..."
Abstract

Cited by 27 (3 self)
 Add to MetaCart
This memo provides information for the Internet community. It does not specify an Internet standard of any kind. Distribution of this memo is unlimited. IESG Note This document is not an IETF Internet Standard. It represents the individual opinion(s) of one or more members of the TMRG Research
A Mechanism for Congestion Control in Computer Networks
 the Existence of Equilibria in Noncooperative Optimal Flow Control&quot;, Proceedings of ITC Workshop
, 1989
"... this paper, we present an economic model of a computer network. We then make some simplifying assumptions to derive a mechanism for congestion control. Subsequently we relax some of the assumptions and generalize the solution. We will not present details of implementation of the scheme in reallife ..."
Abstract

Cited by 3 (0 self)
 Add to MetaCart
this paper, we present an economic model of a computer network. We then make some simplifying assumptions to derive a mechanism for congestion control. Subsequently we relax some of the assumptions and generalize the solution. We will not present details of implementation of the scheme in reallife networks: these will be discussed in a forthcoming thesis.
Efficiency of Oblivious versus NonOblivious Schedulers for Optimistic, RateBased Flow Control
 Proceedings of the 16th Annual ACM Symposium on Principles of Distributed Computing
, 1997
"... Abstract. Two important performance parameters of distributed, ratebased flow control algorithms are their locality and convergence complexity. The former is characterized by the amount of global knowledge that is available to their scheduling mechanisms, while the latter is defined as the number o ..."
Abstract

Cited by 3 (2 self)
 Add to MetaCart
Abstract. Two important performance parameters of distributed, ratebased flow control algorithms are their locality and convergence complexity. The former is characterized by the amount of global knowledge that is available to their scheduling mechanisms, while the latter is defined as the number of update operations performed on rates of individual sessions until maxmin fairness is reached. Optimistic algorithms allow any session to intermediately receive a rate larger than its maxmin fair rate; bottleneck algorithms finalize the rate of a session only if it is restricted by a certain, highly congested link of the network. In this work, we present a comprehensive collection of lower and upper bounds on convergence complexity, under varying degrees of locality, for optimistic, bottleneck, ratebased flow control algorithms. Say that an algorithm is oblivious if its scheduling mechanism uses no information of either the session rates or the network topology. We present a novel, combinatorial construction of a capacitated network, which we use to establish a fundamental lower bound of dn n + on the convergence 4 2 complexity of any oblivious algorithm, where n is the number of sessions laid out on a network, and
MaxMin Fair Flow Control Sensitive to Priorities
 Proceedings of the 2nd International Conference on Principles of Distributed Systems
, 1999
"... Flow control is the dominant technique currently used in communication networks for preventing excess traffic from "flooding" the network, and for handling congestion. In ratebased flow control, transmission rates of sessions are adjusted in an endtoend manner through a sequence of opera ..."
Abstract

Cited by 3 (1 self)
 Add to MetaCart
Flow control is the dominant technique currently used in communication networks for preventing excess traffic from "flooding" the network, and for handling congestion. In ratebased flow control, transmission rates of sessions are adjusted in an endtoend manner through a sequence of operations, each preserving the capacity constraint at each individual network link. In this work, we present a theory of maxmin fair, ratebased flow control sensitive to priorities of different sessions, as a significant extension of the classical theory of maxmin fair, ratebased flow control to the case of networks with guaranteed qualityofservice (QoS). We assume a priority function for each individual session, which maps the session's nominated priority to a transmission rate. We introduce priority maxmin fairness, as a novel and motivatable fairness condition which requires that assigned rates correspond, through the priority functions, to priorities comprising a maxmin vector. A c...
A Packet DroppingBased Incentive Mechanism for M/M/1 Queues with Selfish Users
"... Abstract—We study a novel game theoretic incentive mechanism design problem for network congestion control in the context of selfish users sending data through a single storeandforward router (a.k.a. “server ” in this work). The scenario is modeled as an M/M/1 queueing game with each user (a.k.a. “ ..."
Abstract
 Add to MetaCart
Abstract—We study a novel game theoretic incentive mechanism design problem for network congestion control in the context of selfish users sending data through a single storeandforward router (a.k.a. “server ” in this work). The scenario is modeled as an M/M/1 queueing game with each user (a.k.a. “player”) aiming to optimize a tradeoff between throughput and delay in a selfish distributed manner. We first show that the original game has an inefficient unique Nash Equilibrium (NE). In order to improve the outcome efficiency, we propose an incentivizing packet dropping scheme that can be easily implemented at the server. We then show that if the packet dropping scheme is a function of the sum of arrival rates, we have a modified M/M/1 queueing game that is an ordinal potential game with a unique NE. In particular, for a linear packet dropping scheme, which is similar to the Random Early Detection (RED) algorithm used with TCP, we show that there exists a unique Nash Equilibrium. For this scheme, the social welfare (expressed either as the summation of utilities of all players or log summation of utilities of all players) at the equilibrium point can be arbitrarily close to the social welfare at the global optimal point. Finally, we show that the simple best response dynamic converges to this unique efficient Nash Equilibrium. I.