Performance Analysis of Feedback Controlled Data Packet Transmission over High-Speed Networks (original) (raw)
Related papers
Analysis of discarding policies in high-speed networks
IEEE Journal on Selected Areas in Communications, 1998
Networked applications generate messages that are segmented into smaller, fixed or variable size packets, before they are sent through the network. In high-speed networks, acknowledging individual packets is impractical; so when congestion builds up and packets have to be dropped, entire messages are lost. For a message to be useful, all packets comprising it must arrive successfully at the destination. The problem is therefore which packets to discard so that as many complete messages are delivered, and so that congestion is alleviated or avoided altogether.
A Dynamic Rate Control Mechanism for Source Coded Traffic in a Fast Packet Network
IEEE Journal on Selected Areas in Communications, 1991
To achieve better statistical gain for voice and video traffic and to relieve congestion in fast packet networks, a dynamic rate control mechanism is proposed. An analytical model is developed to evaluate the performance of this control mechanism for voice traffic. The feedback delay for the source node to obtain the network congestion information is represented in the model. The study indicates that significant improvement in statistical gain can be realized for smaller capacity links (e.g., links that can accommodate less than 24 voice calls) with a reasonable feedback time (about 100 ms). The tradeoff for increasing the statistical gain is temporary degradation of voice quality to a lower rate. It is shown that whether the feedback delay is exponentially distributed or constant does not significantly affect performance in terms of fractional packet loss and average received coding rate. It is also shown that using the number of calls in talkspurt or the packet queue length as measures of congestion provides comparable performance.
On the performance of early packet discard
IEEE Journal on Selected Areas in Communications, 1997
In a previous paper 3], one of the authors, gave a worst-case analysis for the Early Packet Discard (EPD) technique for maintaining packet integrity during overload in ATM switches. This analysis showed that to ensure 100% goodput during overload under worst-case conditions, requires a bu er with enough storage for one maximum length packet from every active virtual circuit. This paper re nes that analysis, using assumptions that are closer to what we expect to see in practice and examines how EPD performs when the bu er is not large enough to achieve 100% goodput. We show that 100% goodput can be achieved with substantially smaller bu ers than predicted by the worst-case analysis, although the required bu er space can be signi cant when the link speed is substantially higher than the rate of the individual virtual circuits. We also show that high goodputs can be achieved with more modest bu er sizes, but that EPD exhibits anomalies with respect to bu er capacity, in that there are situations in which increasing the amount of bu ering can cause the goodput to decrease. These results are validated by comparison with simulation.
Generalized stochastic performance models for loss-based congestion control
2010
In this paper, we propose a generalized framework for modeling the behavior of prominent congestion-control protocols. Specifically, we define a general class of loss-based congestion-control (LB-CC) mechanisms and demonstrate that many variants of TCP, including those being proposed for high-speed networks, belong to this class.
Window-based control in lossy packet networks
This paper addresses the problem of fair allocation of bandwidth resources on lossy channels and heterogeneous networks. It discusses more particularly the ability of window-based congestion control to support non-congestion related losses. We investigate methods for efficient packet loss recovery by retransmission, and builds on explicit congestion control mechanisms to decouple the packet loss detection from the congestion feedback signals. For different retransmission strategies that respectively rely on conventional cumulative acknowledgments or accurate loss monitoring, we show how the principles underlying the TCP retransmission mechanisms have to be adapted in order to take advantage of an explicit congestion control framework. A novel retransmission timer is proposed to deal with multiple losses of data segments and, in consequence, to allow for aggressive reset of the connection recovery timer. It ensures significant benefit from temporary inflation of the send-out window, and hence the fair share of bottleneck bandwidth between loss-prone and lossy connections. Extensive simulations demonstrate the performance of the new loss monitoring and recovery strategies, when used with two distinct explicit congestion control mechanisms. The first one proposes a simple modification of TCP to support explicit congestion control, based on a coarse binary congestion notification from the routers. The second one, introduced in [1], relies on accurate feedback about congestion to compute fine congestion window adjustment. For both congestion control mechanisms, we observe that retransmissions triggered based on a precise monitoring of losses allow for efficient utilization of lossy links, and provide a fair share of the bottleneck bandwidth between heterogeneous connections, even for high loss ratios and bursty loss processes. Explicit congestion control, combined with appropriate error control strategies, can therefore provide a valid solution to reliable and controlled connections over lossy network infrastructures. In addition, our simulations also reveal that triggering retransmissions based on cumulative acknowledgments is only efficient-in terms of bottleneck utilization and fairness-at high loss rates when used in conjunction with an accurate and finely tuned congestion control. Therefore, we finally recommend the implementation of accurate feedback mechanisms either in the routers (about the level of congestion) or at the receivers (about a packet arrival), in order to provide a fair bandwidth allocation in hybrid networks with explicit window-based control.
Improved analysis of early packet discard
Teletraffic Science and Engineering, 1997
In a previous paper, one of the authors, gave a worst-case analysis for the Early Packet Discard technique for maintaining packet integrity during overload in ATM switches. This analysis showed that to ensure 100% goodput during overload under worst-case conditions, requires a bu er with enough storage for one maximum length packet from every active virtual circuit. This paper re nes that analysis, using assumptions that are closer to what we expect to see in practice. Our principal result is that 100% goodput can be achieved with substantially smaller bu ers, although the required bu er space can be signi cant when the link speed is substantially higher than the rate of the individual virtual circuits. These results are validated by comparison with simulation. We also give a simple analysis to determine the amount of bu ering needed to bound the probability of bu er over ow and under ow.
Explicit Window-Based Control in Lossy Packet Networks
2006
This paper addresses the problem,of fair allocation of bandwidth resources on lossy channels and heterogeneous,networks. It discusses more,particularly the ability of window-based,congestion control to support non-congestion related losses . We investigate methods for efficient packet loss recovery by retransmissio n, and builds on explicit congestion control mechanisms to decouple the packet loss detection from,the congestion feedback,signals. For different retransmission strategies that respect ively rely on conventional cumulative acknowledgments or accurate loss monitoring, we show how the principles underlying the TCP retransmission mechanisms,have to be adapted in order to take advantage,of an explicit congestion control framework. A novel retransmission timer is proposed to deal with multiple losses of data segments and, in consequence, to allow for aggressive reset of the connection recovery timer. It ensures significant ben efit from temporary inflation of the send-out wind...
2008
Network management techniques have long been of interest to the networking research community. The queue size plays a critical role for the network performance. The adequate size of the queue maintains Quality of Service (QoS) requirements within limited network capacity for as many users as possible. The appropriate estimation of the queuing model parameters is crucial for both initial size estimation and during the process of resource allocation. The accurate resource allocation model for the management system increases the network utilization. The present paper demonstrates the results of empirical observation of memory allocation for packet-based services.
TCP performance over end-to-end rate control and stochastic available capacity
IEEE/ACM Transactions on Networking, 2001
Motivated by TCP over end-to-end ABR, we study the performance of adaptive window congestion control, when it operates over an explicit feedback rate-control mechanism, in a situation in which the bandwidth available to the elastic traffic is stochastically time varying. It is assumed that the sender and receiver of the adaptive window protocol are colocated with the rate-control endpoints. The objective of the study is to understand if the interaction of the rate-control loop and the window-control loop is beneficial for end-to-end throughput, and how the parameters of the problem (propagation delay, bottleneck buffers, and rate of variation of the available bottleneck bandwidth) affect the performance. The available bottleneck bandwidth is modeled as a two-state Markov chain. We develop an analysis that explicitly models the bottleneck buffers, the delayed explicit rate feedback, and TCP's adaptive window mechanism. The analysis, however, applies only when the variations in the available bandwidth occur over periods larger than the round-trip delay. For fast variations of the bottleneck bandwidth, we provide results from a simulation on a TCP testbed that uses Linux TCP code, and a simulation/emulation of the network model inside the Linux kernel. We find that, over end-to-end ABR, the performance of TCP improves significantly if the network bottleneck bandwidth variations are slow as compared to the round-trip propagation delay. Further, we find that TCP over ABR is relatively insensitive to bottleneck buffer size. These results are for a short-term average link capacity feedback at the ABR level (INSTCAP). We use the testbed to study EFFCAP feedback, which is motivated by the notion of the effective capacity of the bottleneck link. We find that EFFCAP feedback is adaptive to the rate of bandwidth variations at the bottleneck link, and thus yields good performance (as compared to INSTCAP) over a wide range of the rate of bottleneck bandwidth variation. Finally, we study if TCP over ABR, with EFFCAP feedback, provides throughput fairness even if the connections have different round-trip propagation delays.