PAQM: an adaptive and proactive queue management for end‐to‐end TCP congestion control

Two functions, the congestion indicator (i.e. how to detect congestion) and the congestion control function (i.e. how to avoid and control congestion), are used at a router to support end‐to‐end congestion control in the Internet. Random early detection (RED) (IEEE/ACM Trans. Networking 1993; 1 (4):397–413) enhanced the two functions by introducing queue length averaging and probabilistic early packet dropping. In particular, RED uses an exponentially weighted moving average (EWMA) queue length not only to detect incipient congestion but also to smooth the bursty incoming traffic and its resulting transient congestion. Following RED, many active queue management (AQM)‐based extensions have been proposed. However, many AQM proposals have shown severe problems with detection and control of the incipient congestion adaptively to the dynamically changing network situations. In this paper, we introduce and analyse a feedback control model of TCP/AQM dynamics. Then, we propose the Pro‐active Queue Management (PAQM) mechanism, which is able to provide proactive congestion avoidance and control using an adaptive congestion indicator and a control function under a wide range of traffic environments. The PAQM stabilizes the queue length around the desired level while giving smooth and low packet loss rates and high network resource utilization. Copyright © 2004 John Wiley & Sons, Ltd.     

Modified RED gateways under bursty traffic

We propose modified random early detection (MRED) gateways for congestion avoidance in TCP/IP networks. MRED aims at providing better control over the burstiness level while remaining the advantages of RED. We use ns2 to simulate a series of network configurations and the numerical results demonstrate that MRED can achieve higher link utilization compared to RED without sacrificing its advantages in all simulation scenarios. The improvement becomes more significant under bursty traffic.     

Performance evaluation of high speed TCP over optical burst switching networks

Study of TCP performance over OBS networks has been an important problem of research lately and it was found that due to the congestion control mechanism of TCP and the inherent bursty losses in the Optical Burst Switching (OBS) network, the throughput of TCP connections degrade. On the other hand, High Speed TCP (HSTCP) was proposed as an alternative to the use of TCP in high bandwidth-delay product networks. HSTCP aggressively increases the congestion window used in TCP, when the available bandwidth is high and decreases the window cautiously in response to a congestion event. In this work, we make a thorough simulation study of HSTCP over OBS networks. While the earlier works in the literature used a linear chain of nodes as the network topology for the simulation, we use the popular 14-node NSFNET topology that represents an arbitrary mesh network in our study. We also study the performance of HSTCP over OBS for different bandwidths of access networks. We use two different cases for simulations where in the first HSTCP connections are routed on disjoint paths while in the second they contend for resources in the network links. These cases of simulations along with the mesh topology help us clearly distinguish between the congestion and contention losses in the OBS network and their effect on HSTCP throughput. For completeness of study, we also simulate TCP traffic over OBS networks in all these cases and compare its throughput with that of HSTCP. We observe that irrespective of the access network bandwidth and the burst loss rate in the network, HSTCP outperforms TCP in terms of the throughput and robustness against multiple burst losses up to the expected theoretical burst loss probability of 10⁻³.     

Buffer and bandwidth allocation for TCP data traffic: experimental results

This paper discusses the dimensioning of buffers and the bandwidth allocation for data traffic in the ATM network. Data traffic is notoriously complex and bursty, making such dimensioning a difficult task. However, the COMBINE project, when dimensioning their InterWorking Units (IWUs), adopted a Poissonian packet arrival model, based upon the argument that burstiness at timescales higher than that of a packet arrival are a problem to be tackled by flow control at higher layers. This paper presents experimental results from the COMBINE testbed that show that this hypothesis was justified and that good TCP goodput was obtained based upon this dimensioning approach, due to TCP's ability to adapt to network congestion. However, it is also shown that it was the TCP algorithm that was ultimately responsible for controlling the packet loss ratio in the network and not the bandwidth allocation or buffer size. The results highlight the importance of taking into account the mutual influence between the ATM layer and the transport layer congestion control algorithms.     

前向主动网络拥塞控制算法及其性能分析

本文提出了一种基于主动式网络(Active Networks)技术的拥塞控制算法FACC(Forward Active Networks Congestion Control).与传统的TCP(Transport Control Protocol)相比,FACC算法通过在网络结点直接提供拥塞检测和拥塞控制机制,大大缩短源端点的拥塞反应时间,从本质上提高了网络拥塞检测和控制的性能,从而提高了终端用户的平均吞吐量.文中还利用计算机仿真研究了FACC算法在各种网络条件下的性能,并与传统的Tahoe,Reno,NewReno及SACK TCP协议做了对比.结果表明无论网络中存不存在非受控数据流时,FACC控制算法均能明显地提高用户终端的平均吞吐量,并且由于采用FACC控制算法而增加的网络结点运算迟延也很小.     

FlowMate: scalable on-line flow clustering

We design and implement an efficient on-line approach, FlowMate, for clustering flows (connections) emanating from a busy server, according to shared bottlenecks. Clusters can be periodically input to load balancing, congestion coordination, aggregation, admission control, or pricing modules. FlowMate uses in-band (passive) end-to-end delay measurements to infer shared bottlenecks. Delay information is piggybacked on feedback from the receivers, or, if impossible, TCP or application round-trip time estimates are used. We simulate FlowMate and examine the effects of network load, traffic burstiness, network buffer sizes, and packet drop policies on clustering correctness, evaluated via a novel accuracy metric. We find that coordinated congestion management techniques are more fair when integrated with FlowMate. We also implement FlowMate in the Linux kernel v2.4.17 and evaluate its performance on the Emulab testbed, using both synthetic and tcplib-generated traffic. Our results demonstrate that clustering of medium to long-lived flows is accurate, even with bursty background traffic. Finally, we validate our results on the Internet Planetlab testbed.     

A stochastic dynamic controller for random early detection queues using a Kalman filter algorithm

The stabilizing random early detection (RED) congestion control algorithm in transmission control protocol (TCP)/IP networks is a control theory problem. Significant attention has been drawn to this problem in the networking and control theory research communities. In this paper, we use a nonlinear dynamic model of the TCP RED congestion control algorithm to analyze and design active queue management (AQM) control systems. A linearized model of RED behavior around its nominal operating point which implicitly includes the delay in the control signal is derived. It is assumed that the system model is corrupted at the input and output by zero mean white Gaussian noise signals. An optimal state feedback stochastic controller is designed for the linearized model of the system in conjunction with a Kalman filter for state estimation. To illustrate the proposed design methodology, simulations results are presented and discussed. The proposed stochastic controller is applied to the nonlinear model of the system; Simulation results indicate that the proposed controller keeps the queue length bounded in an appropriate stochastic sense. Copyright © 2009 John Wiley & Sons, Ltd.     

Accumulation-based congestion control

This paper generalizes the TCP Vegas congestion avoidance mechanism and uses accumulation , buffered packets of a flow inside network routers, as a congestion measure based on which a family of congestion control schemes can be derived. We call this model Accumulation-based Congestion Control (ACC), which fits into the nonlinear optimization framework proposed by Kelly. The ACC model serves as a reference for packet-switching network implementations. We show that TCP Vegas is one possible scheme under this model. It is well known that Vegas suffers from round trip propagation delay estimation error and reverse path congestion. We therefore design a new Monaco scheme that solves these problems by employing an out-of-band, receiver-based accumulation estimator, with the support of two FIFO priority queues from the (congested) routers. Comparisons between these two schemes demonstrate that Monaco does not suffer from the problems mentioned above and achieves better performance than Vegas. We use ns-2 simulations and Linux implementation experiments to show that the static and dynamic performance of Monaco matches the theoretic results. One key issue regarding the ACC model in general, i.e., the scalability of bottleneck buffer requirement, and a solution using a virtual queueing algorithm are discussed and evaluated.     

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     

联合无线网络编码与TCP Vegas的多播协议优化研究

为解决移动自组网中网络编码多播路由协议因业务传输负载增大,而产生的网络拥塞现象,本文提出了一种可靠的基于TCP Vegas窗口拥塞控制的网络编码多播路由协议。该协议的核心思想是发送节点采用发送窗口自调整和反馈消息触发发送窗口调整的机制,综合的调节数据包的发送速率,来改善网络拥塞现象,从而可以降低丢包率。仿真结果表明,当传输负载增大时,基于窗口拥塞控制的网络编码多播路由协议可使得系统的总开销大大降低,分组投递率获得了相对的提升。     

A stochastic model for the behavior of multiple TCP NewReno sources over optical burst switching network

This work proposes a stochastic model to characterize the transmission control protocol (TCP) over optical burst switching (OBS) networks which helps to understand the interaction between the congestion control mechanism of TCP and the characteristic bursty losses in the OBS network. We derive the steady-state throughput of a TCP NewReno source by modeling it as a Markov chain and the OBS network as an open queueing network with rejection blocking. We model all the phases in the evolution of TCP congestion window and evaluate the number of packets sent and time spent in different states of TCP. We model the mixed assembly process, burst assembler and disassembler modules, and the core network using queueing theory and compute the burst loss probability and end-to-end delay in the network. We derive expression for the throughput of a TCP source by solving the models developed for the source and the network with a set of fixed-point equations. To evaluate the impact of a burst loss on each TCP flow accurately, we define the burst as a composition of per-flow-bursts (which is a burst of packets from a single source). Analytical and simulation results validate the model and highlight the importance of accounting for individual phases in the evolution of TCP congestion window.     

A Unified Approach to Congestion Control and Node-Based Multipath Routing

The paper considers a TCP/IP-style network with flow control at end-systems based on congestion feedback and routing decisions at network nodes on a per-destination basis. The main generalization with respect to standard IP is to allow routers to split their traffic in a controlled way between the outgoing links. We formulate global optimization criteria, combining those used in the congestion control and traffic engineering, and propose decentralized controllers at sources and routers to reach these optimal points, based on congestion price feedback. We first consider adapting the traffic splits at routers to follow the negative price gradient; we prove this is globally stabilizing when combined with primal congestion control, but can exhibit oscillations in the case of dual congestion control. We then propose an alternative anticipatory control of routing, proving its stability for the case of dual congestion control. We present a concrete implementation of such algorithms, based on queueing delay as congestion price. We use TCP-FAST for congestion control and develop a multipath variant of the distance vector routing protocol RIP. We demonstrate through ns2-simulations the collective behavior of the system, in particular that it reaches the desired equilibrium points.     

Steady state and transient state behaviours analyses of TCP connections considering interactions between TCP connections and network

The Internet uses a window‐based congestion control mechanism in transmission control protocol (TCP). In the literature, there have been a great number of analytical studies on TCP. Most of those studies have focused on the statistical behaviour of TCP by assuming a constant packet loss probability in the network. However, the packet loss probability, in reality, changes according to the packet transmission rates from TCP connections. Conversely, the window size of a TCP connection is dependent on the packet loss probability in the network. In this paper, we explicitly model the interaction between the congestion control mechanism of TCP and the network as a feedback system. By using this model, we analyse the steady state and the transient state behaviours of TCP. We derive the throughput and the packet loss probability of TCP, and the number of packets queued in the bottleneck router. We then analyse the transient state behaviour using a control theoretic approach, showing the influence of the number of TCP connections and the propagation delay on the transient state behaviour of TCP. Copyright © 2005 John Wiley & Sons, Ltd.     

Rate-adaptive snoop: a TCP enhancement scheme over rate-controlled lossy links

We study TCP performance over the wireless links deploying a wireless rate-control technique, whose link characteristics are identified by variable link rate and bursty transmission error. We present a TCP enhancement scheme, called rate-adaptive snoop (RA-Snoop). RA-Snoop caches TCP packets selectively based on the wireless channel condition and the cached packets are retransmitted locally over the wireless link in case corruption loss is detected. In addition, for effective adaptation to variable bandwidth, RA-Snoop calculates the window feedback based on the bandwidth-delay product estimation and the queue level, then conveys this feedback information on the receiver's advertised window field in the acknowledgements returning to TCP sources. We compare the performance of RA-Snoop with that of existing schemes in the aspect of goodput and fairness. Results from simulations reveal that RA-Snoop achieves significant improvements over the existing schemes for various traffic scenarios.     

Improving Transport Layer Performance in Multihop Ad Hoc Networks by Exploiting MAC Layer Information

The traditional TCP congestion control mechanism encounters a number of new problems and suffers a poor performance when the IEEE 802.11 MAC protocol is used in multihop ad hoc networks. Many of the problems result from medium contention at the MAC layer. In this paper, we first illustrate that severe medium contention and congestion are intimately coupled, and TCP's congestion control algorithm becomes too coarse in its granularity, causing throughput instability and excessively long delay. Further, we illustrate TCP's severe unfairness problem due to the medium contention and the tradeoff between aggregate throughput and fairness. Then, based on the novel use of channel busyness ratio, a more accurate metric to characterize the network utilization and congestion status, we propose a new wireless congestion control protocol (WCCP) to efficiently and fairly support the transport service in multihop ad hoc networks. In this protocol, each forwarding node along a traffic flow exercises the inter-node and intra-node fair resource allocation and determines the MAC layer feedback accordingly. The end-to-end feedback, which is ultimately determined by the bottleneck node along the flow, is carried back to the source to control its sending rate. Extensive simulations show that WCCP significantly outperforms traditional TCP in terms of channel utilization, delay, and fairness, and eliminates the starvation problem     

iTCP: an intelligent TCP with neural network based end-to-end congestion control for ad-hoc multi-hop wireless mesh networks

Maintaining the performance of reliable transport protocols, such as transmission control protocol (TCP), over wireless mesh networks (WMNs) is a challenging problem due to the unique characteristics of data transmission over WMNs. The unique characteristics include multi-hop communication over lossy and non-deterministic wireless mediums, data transmission in the absence of a base station, similar traffic patterns over neighboring mesh nodes, etc. One of the reasons for the poor performance of conventional TCP variants over WMNs is that the congestion control mechanisms in conventional TCP variants do not explicitly account for these unique characteristics. To address this problem, this paper proposes a novel artificial intelligence based congestion control technique for reliable data transfer over WMNs. The synergy with artificial intelligence is established by exploiting a carefully designed neural network (NN) in the congestion control mechanism. We analyze the proposed NN based congestion control technique in detail and incorporate it into TCP to create a new variant that we name as intelligent TCP or iTCP. We evaluate the performance of iTCP using both ns-2 simulations and real testbed experiments. Our evaluation results demonstrate that our proposed congestion control technique exhibits a significant improvement in total network throughput and average energy consumption per transmitted bit compared to the congestion control techniques used in other TCP variants.     

Self-similar traffic and network dynamics

One of the most significant findings of traffic measurement studies over the last decade has been the observed self-similarity in packet network traffic. Subsequent research has focused on the origins of this self-similarity, and the network engineering significance of this phenomenon. This paper reviews what is currently known about network traffic self-similarity and its significance. We then consider a matter of current research, namely, the manner in which network dynamics (specifically, the dynamics of transmission control protocol (TCP), the predominant transport protocol used in today's Internet) can affect the observed self-similarity. To this end, we first discuss some of the pitfalls associated with applying traditional performance evaluation techniques to highly-interacting, large-scale networks such as the Internet. We then present one promising approach based on chaotic maps to capture and model the dynamics of TCP-type feedback control in such networks. Not only can appropriately chosen chaotic map models capture a range of realistic source characteristics, but by coupling these to network state equations, one can study the effects of network dynamics on the observed scaling behavior We consider several aspects of TCP feedback, and illustrate by examples that while TCP-type feedback can modify the self-similar scaling behavior of network traffic, it neither generates it nor eliminates it     

TCP-LP: low-priority service via end-point congestion control

Service prioritization among different traffic classes is an important goal for the Internet. Conventional approaches to solving this problem consider the existing best-effort class as the low-priority class, and attempt to develop mechanisms that provide "better-than-best-effort" service. In this paper, we explore the opposite approach, and devise a new distributed algorithm to realize a low-priority service (as compared to the existing best effort) from the network endpoints. To this end, we develop TCP Low Priority (TCP-LP), a distributed algorithm whose goal is to utilize only the excess network bandwidth as compared to the "fair share" of bandwidth as targeted by TCP. The key mechanisms unique to TCP-LP congestion control are the use of one-way packet delays for early congestion indications and a TCP-transparent congestion avoidance policy. The results of our simulation and Internet experiments show that: 1) TCP-LP is largely non-intrusive to TCP traffic; 2) both single and aggregate TCP-LP flows are able to successfully utilize excess network bandwidth; moreover, multiple TCP-LP flows share excess bandwidth fairly; 3) substantial amounts of excess bandwidth are available to the low-priority class, even in the presence of "greedy" TCP flows; 4) the response times of web connections in the best-effort class decrease by up to 90% when long-lived bulk data transfers use TCP-LP rather than TCP; 5) despite their low-priority nature, TCP-LP flows are able to utilize significant amounts of available bandwidth in a wide-area network environment.     

Impact of Doppler spread and adaptive modulation on TCP throughput in Rayleigh fading channels

A number of wireless systems have recently adopted adaptive modulation (AM) schemes to improve its efficiency. In this letter, our aim is to study the impact Doppler spread and adaptive modulation has on transmission control protocol (TCP) throughput in Rayleigh fading channels. We consider a finite state Markov channel (FSMC) model, which is a useful model for analyzing radio channel with nonindependent fading. Furthermore, we use a Markov model for TCP evolution and evaluate the TCP performance by computer simulations. In our simulations we have compared the TCP Reno scheme with TCP Tahoe scheme. The results indicate that a large Doppler spread leads to lower TCP throughput due to more frequent transitions of channel states and modulation schemes which make it difficult for the TCP congestion control mechanism to accommodate the dynamic link characteristics.