Random Early Detection Gateway Research Articles

RED-Based Technique for Detecting and Avoiding Anomaly Network Congestion

Active Queue Management (AQM) is an intelligent drop of network packets within the network routers as a result of buffer saturation to reduce network congestion. Random Early Detection (RED) mechanism is the most used mechanism among the existing AQMs, as it can avoid network congestion at the early stages. The RED mechanism prompts users to reduce their windows size when the queue average has exceeded a predefined threshold. However, some users do not respond to these notifications, and therefore, RED drops all packets in the link. This will generate a false alarm because the well-behaving packets on that link will be dropped as well. This study proposes a technique for monitoring gateways’ queues and dropping only the misbehaving packets. Every RED gateway reports the packet transfer rate (PTR) of misbehaving users that have exceeded their bandwidth shares to the SLA management (SLAM) unit. The SLAM unit will compute the users’ PTRs and compare them with their predefined ratios in the SLA. Users that have exceeded the SLA-predefined threshold will be considered as misbehaving and isolated from the normal users. The obtained results show that the proposed technique is efficient in identifying the misbehaving packets and avoiding network congestion.

Nonlinear Dynamics in High-speed Wireless Networks Congestion Control Model with TCP LogWestwood+ under RED

In this paper, we proposed a first-order discrete time nonlinear dynamic model of congestion control system with TCP LogWestwood+ (TCPLog) connections and random early detection (RED) gateway in high-speed wireless networks. The model is used to analyze nonlinear dynamics of the TCPLog/RED network and its stability with respect to various RED controller and system parameters. Bifurcation and chaos behaviors are shown to occur when parameters varied. By theoretical derivation, the fixed point and the critical value are obtained, and the nature of the bifurcation is determined. Furthermore, bifurcation diagrams and Lyapunov exponent are exploited to verify the theoretical results. Finally, the bifurcation and chaotic phenomena of the congestion control system are numerically studied with TCPLog connections and RED gateway

Service Violation Monitoring Model for Detecting and Tracing Bandwidth Abuse

Bandwidth abuse is a critical Internet service violation. However, its origins are difficult to detect and trace given similarities between abusive and normal traffic. So far, there is no capable and scalable mechanism to deal with bandwidth abuse. This paper proposes a distributed edge-to-edge model for monitoring service level agreement (SLA) violations and tracing abusive traffic to its origins. The mechanism of policing misbehaving user traffic at a single random early detection (RED) gateway is used in the distributed monitoring of SLA violations, including violations carried out through several gateways. Each RED gateway reports misbehaving users who have been sent notifications of traffic policing to an SLA monitoring unit. Misbehaving users are considered suspicious users and their consumed bandwidth shares are aggregated at every gateway to be compared with SLA-specified ratios. Bandwidth is abused when SLA-specified ratios are exceeded. By reporting bandwidth abuse, illegitimate users can be isolated from legitimate ones and source hosts of abusive traffic may be traced. Approximate simulation results show that the proposed model can detect any SLA violation and identify abusive users. In addition, the proposed model can trace user violations back to their source machines in real time.

CARED: Cautious Adaptive RED gateways for TCP/IP networks

Random Early Detection (RED) is a widely deployed active queue management algorithm that improves the overall performance of the network in terms of throughput and delay. The effectiveness of RED algorithm, however, highly depends on appropriate setting of its parameters. Moreover, the performance of RED is quite sensitive to abrupt changes in the traffic load. In this paper, we propose a Cautious Adaptive Random Early Detection (CARED) algorithm that dynamically varies maximum drop probability based on the level of traffic load to improve the overall performance of the network. Based on extensive simulations conducted using Network Simulator-2 (ns-2), we show that CARED algorithm reduces the packet drop rate and achieves high throughput as compared to RED, Adaptive RED and Refined Adaptive RED. Unlike other RED based algorithms, CARED algorithm does not introduce new parameters to achieve performance gain and hence can be deployed without any additional complexity.

Adding Randomness to Modeling Internet TCP-RED Systems With Interactive Gateways

In Internet gateways with active queue management, effective algorithms are often used to enhance the traffic flows through the gateway. The random early detection (RED) algorithm is a widely used algorithm for controlling the transmission control protocol (TCP) flows. Specifically, the RED algorithm has deliberately introduced randomness into the explicit congestion notification (ECN) marking mechanism in order to enhance the fairness and to avoid synchronization of TCP flows through the gateway. As randomness plays a key role in providing better performance for RED gateways than the DropTail gateways, knowledge of the effect of the random mechanism is important to the understanding the behavior of the system. The fluid flow model (FFM), ignoring the randomness of ECN markings, has provided a fast alternative to the slow ns-2 network simulation tool in studying the TCP flow in the RED gateway. In this brief, the random ECN marking mechanism in RED gateways has been studied. The randomness of the RED ECN marking algorithm is implemented into the FFM. Verified using ns-2 simulations, the proposed model shows better dynamical performance than FFM in both the single-bottleneck dumbbell network and in networks with interactive bottlenecks.

Nonlinear dynamics in Internet congestion control model with TCP Westwood under RED

This article develops a discrete time dynamic feedback model of a congestion control system for a simple network with TCP Westwood (TCPW) connections and a single bottleneck link with random early detection (RED) gateway. By using this model, the nonlinear dynamics of the TCPW/RED network are analyzed and its parameter sensitivities are studied. It is shown that periodic doubling bifurcation occurs when the RED control parameters or other parameters are varied. By theoretical analysis, the fixed point, the critical value of parameters and the nature of the bifurcation are determined. Moreover, by using bifurcation diagrams and Lyapunov exponent, the result of theoretical analysis is validated and the bifurcation and chaotic phenomena are numerically studied of the congestion control system with TCPW connections and RED gateway.

OSCILLATION AND PERIOD DOUBLING IN TCP/RED SYSTEM: ANALYSIS AND VERIFICATION

It has been known that a bottleneck RED (Random Early Detection) gateway can become oscillatory when regulating multiple identical TCP (Transmission Control Protocol) flows. However, a systematic explanation of such oscillatory behavior is not available. In this paper, we first use the fluid-flow model to derive the system characteristic frequency, and then compare with the frequencies of the RED queue length waveforms observed from "ns-2" simulations. The "ns-2" simulator is the only viable simulation tool accepted by industry for verification purposes. Analysis of the TCP source frequency distribution reveals the occurrence of period doubling when the system enters the instability region as the filter resolution varies. Since random events and a large number of TCP flows are involved in the process of generating the average system dynamics, a statistical viewpoint is taken in the analysis. Our results reflect the true system behavior as they are based on data from "ns-2" simulations rather than numerical simulations of analytical models. The physical mechanism of oscillation is explained in terms of the difference in the TCP source frequency and the TCP-RED system characteristic frequency.

A modified random early detection algorithm: Fuzzy logic based approach

In this paper, a fuzzy logic implementation of the random early detection (RED) mechanism [1] is presented. The main objective of the proposed fuzzy controller is to reduce the loss probability of the RED mechanism without any change in channel utilization. Based on previous studies, it is clear that the performance of RED algorithm is extremely related to the traffic load as well as to its parameters setting. Using fuzzy logic capabilities, we try to dynamically tune the loss probability of the RED gateway. To achieve this goal, a two-input-single-output fuzzy controller is used. To achieve a low packet loss probability, the proposed fuzzy controller is responsible to control the max p parameter of the RED gateway. The inputs of the proposed fuzzy controller are 1) the difference between average queue size and a target point, and 2) the difference between the estimated value of incoming data rate and the target link capacity. To evaluate the performance of the proposed fuzzy mechanism, several trials with file transfer protocol (FTP) and burst traffic were performed. In this study, the ns-2 simulator [2] has been used to generate the experimental data. All simulation results indicate that the proposed fuzzy mechanism outperforms remarkably both the traditional RED and Adaptive RED (ARED) mechanisms [3]–[5].

A review of central issues - Signal processing challenges in active queue management

The publication of Floyd and Jacobson's seminal paper "Random early detection gateways for congestion avoidance" (1993), marked a new direction in networking research and began what is perhaps the most investigated example of cross-layer optimization. While this paper has inspired an immense amount of work in research, many open problems in active queue management (AQM) still remain. This article seeks to frame these problems in terms accessible to the signal processing researchers. The basic idea of AQM has been provided as well as its objective and overviews of a sample of different approaches. The signal processing aspects of the AQM are discussed, specifically the problem of predicting congestion, approaches to detecting changes in network traffic, an estimation problem, dithering, and quantization.

Instability of a Tandem Network and Its Propagation Under RED

The interaction between a random early detection (RED) gateway and transmission control protocol (TCP) connections has been shown to lead to a rich set of nonlinear phenomena in single bottleneck cases. We extend this work and study the interaction of TCP connections with RED gateways in a simple tandem network, using a nonlinear first-order discrete-time model. We demonstrate that the nonlinear behavior of TCP can result in both smooth and nonsmooth bifurcations, leading to chaos. We show that the instabilities can be induced at both bottlenecks by changing the system parameters only at one of the bottlenecks while fixing the parameters at the other, thus demonstrating the propagation of instability. Moreover, we show that locally sufficient conditions for stability based on single node analysis are not sufficient for global network stability.

RED gateway congestion control using median queue size estimates

This paper focuses on the queue size estimation problem in random early detection (RED) gateways. Queue size estimation plays a fundamental role in the congestion control dynamics of RED, as it determines gateways' awareness of network congestion, which in turn determines the packet dropping/marking decision. Conventional RED gateways use exponentially weighted moving averages (EWMA) to estimate the queue size. These infinite impulse response (IIR) filters require very small EWMA weights in order to effectively avoid nonlinear instabilities in RED and to filter out bursty increases in the queue size. While small EWMA weights enable gateways to accommodate transient congestion, they also lead to gateways' failure to closely track rapid queue size depletion and thus causes link under utilization. We investigate the use of simple nonlinear queue size estimators. In particular, we study the congestion control dynamics of a network where adaptive weighted median filters are used for queue size estimation by the gateways. Analytical results for the expected queue size in the steady state are derived. Under this new queue size estimation framework, design guidelines for the remaining RED parameters are provided. Simulation results show that the proposed algorithm provides greater network power, better prevention of global synchronization, and a fairer treatment to bursty traffic than the RED algorithm does.

A comparison of RED’s byte and packet modes

Routers making use of random early detection (RED) queueing take action to notify sources of growing congestion levels in the network before their resources are exhausted. The RED system hinges on two calculations: tracking the average queue size and the probability that an incoming packet is marked for congestion. These two calculations can be done in terms of the number of packets arriving at the router or in terms of the size of those packets (in bytes). Intuitively, these calculation methods offer different costs and benefits to traffic. This paper quantitatively assesses the impact of using the different queueing and marking methods on the performance of traffic traversing a RED gateway. We show that in some cases the calculation method makes a difference in the performance of the system, while in other cases the choice has little impact. We also provide a framework for rating the RED variants in particular situations in an attempt to aid in the choice of variant to use in a specific situations.

Low pass filter/over drop avoidance (LPF/ODA): an algorithm to improve the response time of RED gateways

AbstractThe low pass filter (LPF) algorithm, which is used to calculate the average queue length in the random early detection (RED) algorithm, results in a poor response time when RED recovers from congestion. In this paper, we propose a new algorithm, called low pass filter/over drop avoidance (LPF/ODA), to improve the response time of RED. LPF/ODA is based on reducing the time required by RED to recover from long‐term congestion. Simulation results have been used to show that the proposed LPF/ODA algorithm significantly improves the response time, throughput and delay of RED gateways. Copyright © 2002 John Wiley & Sons, Ltd.

A comparison of RED's byte and packet modes

Routers making use of Random Early Detection (RED) queueing take action to notify sources of growing congestion levels in the network before their resources are exhausted. The RED system hinges on two calculations: tracking the average queue size and the probability that an incoming packet is marked for congestion. These two calculations can be done in terms of the number of packets arriving at the router or in terms of the size of those packets (in bytes). Intuitively, these calculation methods off er different costs and benefits to traffic. This paper quantitatively assesses the impact of using the different queueing and marking methods on the performance of traffic traversing a RED gateway. We show that in some cases the calculation method makes a difference in the performance of the system, while in other cases the choice has little impact. We also provide a framework for rating the RED variants in particular situations in an attempt to aid in the choice of variant to use in a specific situations.

Random early detection gateways for congestion avoidance

The authors present random early detection (RED) gateways for congestion avoidance in packet-switched networks. The gateway detects incipient congestion by computing the average queue size. The gateway could notify connections of congestion either by dropping packets arriving at the gateway or by setting a bit in packet headers. When the average queue size exceeds a present threshold, the gateway drops or marks each arriving packet with a certain probability, where the exact probability is a function of the average queue size. RED gateways keep the average queue size low while allowing occasional bursts of packets in the queue. During congestion, the probability that the gateway notifies a particular connection to reduce its window is roughly proportional to that connection's share of the bandwidth through the gateway. RED gateways are designed to accompany a transport-layer congestion control protocol such as TCP. The RED gateway has no bias against bursty traffic and avoids the global synchronization of many connections decreasing their window at the same time. Simulations of a TCP/IP network are used to illustrate the performance of RED gateways. >

Traffic phase effects in packet-switched gateways

Much of the traffic in existing packet networks is highly periodic, either because of periodic sources (e.g., real time speech or video, rate control) or because window flow control protocols have a periodic cycle equal to the connection roundtrip time (e.g., a network-bandwidth limited TCP bulk data transfer). Control theory suggests that this periodicity can resonate (i.e., have a strong, non-linear interaction) with deterministic estimation or control algorithms in network gateways. 1 In this paper we define the notion of traffic phase in a packet-switched network and describe how phase differences between competing traffic streams can be the dominant factor in relative throughput. Drop Tail gateways in a TCP/IP network with strongly periodic traffic can result in systematic discrimination against some connections. We demonstrate this behavior with both simulations and theoretical analysis. This discrimination can be eliminated with the addition of appropriate randomization to the network. In particular, analysis suggests that simply coding a gateway to drop a random packet from its queue (rather than the tail) on overflow is often sufficient.We do not claim that Random Drop gateways solve all of the problems of Drop Tail gateways. Biases against bursty traffic and long roundtrip time connections are shared by both Drop Tail and Random Drop gateways. Correcting the bursty traffic bias has led us to investigate a different kind of randomized gateway algorithm that operates on the traffic stream, rather than on the queue. Preliminary results show that the Random Early Detection gateway, a newly developed gateway congestion avoidance algorithm, corrects this bias against bursty traffic. The roundtrip time bias (at least in TCP/IP networks) results from the TCP window increase algorithm, not from the gateway dropping policy, and we briefly discuss changes to the window increase algorithm that could eliminate this bias.