Results 1 
4 of
4
A New Approach for Asynchronous Distributed Rate Control of Elastic Sessions in Integrated Packet Networks
 IEEE/ACM Trans. Networking
"... We develop a new class of asynchronous distributed algorithms for the explicit rate control of elastic sessions in an integrated packet network. Sessions can request for minimum guaranteed rate allocations (e.g., MCRs in the ATM context), and, under this constraint, we seek to allocate the maxmin f ..."
Abstract

Cited by 5 (0 self)
 Add to MetaCart
We develop a new class of asynchronous distributed algorithms for the explicit rate control of elastic sessions in an integrated packet network. Sessions can request for minimum guaranteed rate allocations (e.g., MCRs in the ATM context), and, under this constraint, we seek to allocate the maxmin fair rates to the sessions. We capture the integrated network context by permitting the link bandwidths available to elastic sessions to be stochastically time varying. The available capacity of each link is viewed as some statistic of this stochastic process (e.g., a fraction of the mean, or a large deviations Equivalent Service Capacity (ESC)). For fixed available capacity at each link, we show that the vector of maxmin fair rates can be computed from the root of a certain vector equation. A distributed asynchronous stochastic approximation technique is then used to develop a provably convergent distributed algorithm for obtaining the root of the equation, even when the link flows and the ...
A distributed stochastic approximation approach for maxmin fair rate control of flows in packet networks
, 2006
"... We consider a distributed stochastic approximation algorithm that computes maxmin fair rate allocations to several elastic flows sharing a network (an elastic flow is one that can adapt its sending rate to the rate that the network can provide it). The flows are assumed to traverse a fixed sequence ..."
Abstract

Cited by 2 (0 self)
 Add to MetaCart
We consider a distributed stochastic approximation algorithm that computes maxmin fair rate allocations to several elastic flows sharing a network (an elastic flow is one that can adapt its sending rate to the rate that the network can provide it). The flows are assumed to traverse a fixed sequence of links in the network. The available capacities at the network links are modeled as stochastic processes. Each session can request a minimum rate guarantee, hence we work with a notion of maxmin fairness with minimum rates. A major part of this paper is the proof that the rate allocation computed by the stochastic approximation iterations converges to maxmin rate. 1
Internet Packet Transport: Traffic Control and Network Engineering
"... The Internet can be viewed as the transportation infrastructure for the deployment of Information Technology. The flow of traffic in any transportation network needs to be controlled to avoid congestion, to provide special service to flows that need it, and to protect against overloads. In this p ..."
Abstract
 Add to MetaCart
The Internet can be viewed as the transportation infrastructure for the deployment of Information Technology. The flow of traffic in any transportation network needs to be controlled to avoid congestion, to provide special service to flows that need it, and to protect against overloads. In this paper, we consider the control of storeandforward (also called elastic) traffic in the Internet. We first motivate the need for feedback based congestion controls for elastic traffic. These controls provide for fair sharing of the network bandwidth, and hence determine how bandwidth is shared between elastic flows. We discuss two approaches for implementing such controls: in one, the network provides explicit feedbacks to the traffic sources, and in the other the sources adjust their rates based on implicit feedbacks. Then we discuss models that could help in sizing the Internet for guaranteeing minimum throughputs to elastic flows. These models lead to engineering guidelines and su...
A New Approach for Asynchronous Distributed Rate Control of Elastic Sessions in Integrated Packet Networks
"... Abstract — We develop a new class of asynchronous distributed algorithms for the explicit rate control of elastic sessions in an integrated packet network. Sessions can request for minimum guaranteed rate allocations (e.g., MCRs in the ATM context), and, under this constraint, we seek to allocate th ..."
Abstract
 Add to MetaCart
Abstract — We develop a new class of asynchronous distributed algorithms for the explicit rate control of elastic sessions in an integrated packet network. Sessions can request for minimum guaranteed rate allocations (e.g., MCRs in the ATM context), and, under this constraint, we seek to allocate the maxmin fair rates to the sessions. We capture the integrated network context by permitting the link bandwidths available to elastic sessions to be stochastically time varying. The available capacity of each link is viewed as some statistic of this stochastic process (e.g., a fraction of the mean, or a large deviations Equivalent Service Capacity (ESC)). For fixed available capacity at each link, we show that the vector of maxmin fair rates can be computed from the root of a certain vector equation. A distributed asynchronous stochastic approximation technique is then used to develop a provably convergent distributed algorithm for obtaining the root of the equation, even when the link flows and the available capacities are obtained from online measurements. The switch algorithm does not require per connection monitoring, nor does it require per connection marking of control packets. A virtual buffer based approach for online estimation of the ESC is utilised. We also propose techniques for handling large variations in the available capacity owing to the arrivals or departures of CBR/VBR sessions. Finally, simulations results are provided to demonstrate the performance of this class of algorithms in the local and wide area network context. I.