Vegas plus: improving the service fairness

TCP Vegas is a congestion avoidance scheme designed to prevent the periodic packet loss which occurs in traditional schemes. Since Vegas successfully avoids such packet loss, it achieves much higher throughput than TCP Reno. However, it does not concern the fairness among source-destination pairs with different round-trip times (RTTs). We propose a different mechanism to adjust the window size, this allows TCP to provide much better fairness regardless the large variation of RTTs     

TCP CERL: congestion control enhancement over wireless networks

In this paper, we propose and verify a modified version of TCP Reno that we call TCP Congestion Control Enhancement for Random Loss (CERL). We compare the performance of TCP CERL, using simulations conducted in ns-2, to the following other TCP variants: TCP Reno, TCP NewReno, TCP Vegas, TCP WestwoodNR and TCP Veno. TCP CERL is a sender-side modification of TCP Reno. It improves the performance of TCP in wireless networks subject to random losses. It utilizes the RTT measurements made throughout the duration of the connection to estimate the queue length of the link, and then estimates the congestion status. By distinguishing random losses from congestion losses based on a dynamically set threshold value, TCP CERL successfully attacks the well-known performance degradation issue of TCP over channels subject to random losses. Unlike other TCP variants, TCP CERL doesn’t reduce the congestion window and slow start threshold when random loss is detected. It is very simple to implement, yet provides a significant throughput gain over the other TCP variants mentioned above. In single connection tests, TCP CERL achieved an 175, 153, 85, 64 and 88% throughput gain over TCP Reno, TCP NewReno, TCP Vegas, TCP WestwoodNR and TCP Veno, respectively. In tests with multiple coexisting connections, TCP CERL achieved an 211, 226, 123, 70 and 199% throughput improvement over TCP Reno, TCP NewReno, TCP Vegas, TCP WestwoodNR and TCP Veno, respectively.     

Improving Throughput For TCP Vegas

1　Introduction　TransmissionControlProtocol(TCP) [1 ] iswidelyusedinthecurrentInternet,andmanyofpopularInternetservices,includingHyperTextTransferProtocol (HTTP) [2～3] andFileTransferProtocol(FTP) [4] ,useitasthede factostandardtransport layer protocol.TCPVegas[5～6] wasproposedin1 994,whichovercameseveraldrawbacksofTCPReno[7] ,andcanachievebetween 40and 70 percentbetterthroughputascomparedtotheimplementa tionofTCPintheRenodistributionofBSDU NIX[8] andotherTCPversion[9～1 1 ]…     

A remedy for performance degradation of TCP Vegas in asymmetric networks

Many previous studies have indicated that TCP Vegas outperforms TCP Reno. This letter shows that in asymmetric networks in which the bottleneck is on the reverse path rather than on the forward path, Vegas underutilizes the available bandwidth on the forward path by a large margin. A solution that makes use of the TCP timestamp option can effectively restore the throughput on the forward path.     

Fairness and stability of congestion control mechanisms of TCP

In this paper, we focus on fairness and stability of the congestion control mechanisms adopted in several versions of TCP by investigating their time–transient behaviors through an analytic approach. In addition to TCP Tahoe and TCP Reno, we also consider TCP Vegas which has been recently proposed for higher throughput, and enhanced TCP Vegas, which is proposed in this paper for fairness enhancements. We consider the homogeneous case, where two connections have the equivalent propagation delays, and the heterogeneous case, where each connection has different propagation delay. We show that TCP Tahoe and TCP Reno can achieve fairness among connections in the homogeneous case, but cannot in the heterogeneous case. We also show that TCP Vegas can provide almost fair service among connection, but there is some unfairness caused by the essential nature of TCP Vegas. Finally, we explain the effectiveness of our enhanced TCP Vegas in terms of fairness and throughput. This revised version was published online in June 2006 with corrections to the Cover Date.     

Improving TCP Vegas Fairness in Presence of Backward Traffic

TCP Vegas provides better performance compared to traditional TCP Reno schemes. However, backward traffic significantly degrades Vegas efficiency causing underutilization of the available bandwidth and unfairness. In this letter, we present an enhanced congestion control mechanism for TCP Vegas able to correctly remove the undesired impact of reverse traffic on bandwidth sharing. Furthermore, our proposal is easily deployable because it does not require neither clock synchronization nor any special support at the intermediate routers     

TCP-Gvegas with prediction and adaptation in multi-hop ad hoc networks

TCP Vegas performance can be improved since its rate-based congestion control mechanism could proactively avoid possible congestion and packet losses in multi-hop ad hoc networks. Nevertheless, Vegas cannot make full advantage of available bandwidth to transmit packets since incorrect bandwidth estimates may occur due to frequent topology changes caused by node mobility. This paper proposes an improved TCP Vegas based on the grey prediction theory, named TCP-Gvegas, for multi-hop ad hoc networks, which has the capability of prediction and self-adaption, as well as three enhanced aspects in the phase of congestion avoidance. The lower layers’ parameters are considered in the throughput model to improve the accuracy of theoretical throughput. The prediction of future throughput based on grey prediction is used to promote the online control. The optimal exploration method based on Q-Learning and Round Trip Time quantizer are applied to search for the more reasonable changing size of congestion window. Besides, the convergence analysis of grey prediction by using the Lyapunov’s second method proves that a shorter input data length of prediction implies a faster convergence rate. The simulation results show that the TCP-Gvegas achieves a substantially higher throughput and lower delay than Vegas in multi-hop ad hoc networks.     

Analysis of TCP Vegas and TCP Reno

In this paper we use an analytic fluid approach in order to analyze the different features of both Vegas and Reno TCP versions. We then use simulations to confirm our analytic results. When the available bandwidth is high, indeed Vegas can retransmit less than one‐fifth as much data as Reno does, so that the higher the available bandwidth is, the more efficient Vegas is. However, under heavy congestion Vegas behaves like Reno and does not manage to make efficient use of its new mechanism for congestion detection. The analytic results that we obtain are the evolution of the window size, round trip times and their averages, and the average throughput. This revised version was published online in June 2006 with corrections to the Cover Date.     

大带宽时延积网络TCP Vegas自适应慢启动算法

TCP Vegas具有比TCP Reno更好的带宽利用能力和稳定性,但是在带宽时延积较大的网络中,TCP Vegas会出现慢启动过早结束、拥塞窗口过小的问题,降低了传输效率.文中在分析慢启动结束的原因和条件的基础上,提出一种对临时性排队时延进行估计,将其排除后再进行慢启动结束条件判断的TCP Vegas慢启动算法,对不同网络条件有自适应能力.仿真结果表明本算法能有效避免慢启动过早结束,使TCP性能明显改善.     

EVA: a better TCP version for resource‐insufficient networks

TCP Vegas exhibits unfair congestion avoidance mechanism, which aggravates when there are insufficient network resources to accommodate buffer space of a pipe (bandwidth delay product). To remedy this shortcoming, we propose an Enhanced VegAs (EVA) that employs three auxiliary mechanisms: Δ revision, congestion detection and congestion tendency detection. A 2^k factorial design with replications is used to study the effect of the three mechanisms. Our results show that TCP EVA achieves better performance than Vegas under various network conditions. Furthermore, congestion avoidance schemes, such as TCP EVA, perform much better than congestion control schemes, such as TCP Reno, in resource‐insufficient networks. Copyright © 2002 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.     

Performance Analysis of TCP with Delayed Acknowledgments in Multi-hop Ad-hoc Networks

The acknowledgment strategy has great potential to increase TCP throughput when it runs over 802.11 MAC protocol. In particular, TCP acknowledgments (ACK) carry out an extensive number of medium accesses as they compete in the same route as data packets for media. In this paper, we first propose a dynamic TCP-MAC interaction strategy which tries to reduce the number of induced ACKs by monitoring the channel condition. To this end, the total collision probability collected along the path from sender to receiver in MAC layer has been used to properly set the number of delayed ACKs (DA) in TCP. Based on the estimated collision probability, TCP sender dynamically adjusts itself to the channel condition by delaying less ACKs in high traffic conditions and more ACKs in low traffic conditions. The simulation results show a throughput improvement up to 15% over the existing method called Dynamic Adaptive Acknowledgment (TCP-DAA) and much more over the regular TCP in different scenarios dealing with a dynamic loss rate. In addition, we show that our proposed strategy does not always benefit from a fixed delay policy along with a fixed congestion window size. In fact, the optimal number of delayed ACKs is based on the path length of a TCP connection and a large delay window may solely improve TCP throughput in short ranges with less number of flows. However, in a longer path congestion window limit provides more throughput gain.     

Dropping Policy with Burst Retransmission to Improve the Throughput of TCP Over Optical Burst-switched Networks

A major concern in optical burst-switched (OBS) networks is contention, which occurs when more than one bursts contend for the same data channel at the same time. Due to the bufferless nature of OBS networks, these contentions randomly occur at any degree of congestion in the network. When contention occurs at any core node, the core node drops bursts according to its dropping policy. Burst loss in OBS networks significantly degrades the throughput of TCP sources in the local access networks because current TCP congestion control mechanisms perform a slow start phase mainly due to contention rather than heavy congestion. However, there has not been much study about the impact of burst loss on the performance of TCP over OBS networks. To improve TCP throughput over OBS networks, we first introduce a dropping policy with burst retransmission that retransmits the bursts dropped due to contention, at the ingress node. Then, we extend the dropping policy with burst retransmission to drop a burst that has experienced fewer retransmissions in the event of contention at a core node in order to reduce the number of events that a TCP source enters the slow start phase due to contention. In addition, we propose to limit the number of retransmissions of each burst to prevent severe congestion. For the performance evaluation of the proposed schemes, we provide an analytic throughput model of TCP over OBS networks. Through simulations as well as analytic modeling, it is shown that the proposed dropping policy with burst retransmission can improve TCP throughput over OBS networks compared with an existing dropping policy without burst retransmission.     

基于Gilbert丢包机制的TCP吞吐量模型

 丢包机制是推导TCP吞吐量模型的关键,直接影响模型的准确性.本文利用四状态Gilbert丢包机制来描述端到端路径上的丢包行为,对TCP的拥塞控制过程进行建模,在此基础上提出了一种更精确的TCP吞吐量模型.实验表明,改进的模型能较好的与实际值相拟合,可以更精确地预测实际TCP数据流的吞吐量性能.     

TCP Vegas: end to end congestion avoidance on a global Internet

Vegas is an implementation of TCP that achieves between 37 and 71% better throughput on the Internet, with one-fifth to one-half the losses, as compared to the implementation of TCP in the Reno distribution of BSD Unix. This paper motivates and describes the three key techniques employed by Vegas, and presents the results of a comprehensive experimental performance study, using both simulations and measurements on the Internet, of the Vegas and Reno implementations of TCP     

TCP/IP performance with random loss and bidirectional congestion

With the growth in Internet access services over networks with asymmetric links such as asymmetric digital subscriber line (ADSL) and cable-based access networks, it becomes crucial to evaluate the performance of TCP/IP over systems in which the bottleneck link speed on the reverse path is considerably slower than that on the forward path. In this paper, we provide guidelines for designing network control mechanisms for supporting TCP/IP. We determine the throughput as a function of buffering, round-trip times, and normalized asymmetry (defined as the ratio of the transmission time of acknowledgment (ACK) in the reverse path to that of data packets in the forward path). We identify three modes of operation which are dependent on the forward buffer size and the normalized asymmetry, and determine the conditions under which the forward link is fully utilized. We also show that drop-from-front discarding of ACKs on the reverse link provides performance advantages over other drop mechanisms in use. Asymmetry increases the TCP already high sensitivity to random packet losses that occur on a time scale faster than the connection round-trip time. We generalize the by-now well-known relation relating the square root of the random loss probability to obtained TCP throughput, originally derived considering only data path congestion. Specifically, random loss leads to significant throughput deterioration when the product of the loss probability, the normalized asymmetry and the square of the bandwidth delay product is large. Congestion in the reverse path adds considerably to TCP unfairness when multiple connections share the reverse bottleneck link. We show how such problems can be alleviated by per-connection buffer and bandwidth allocation on the reverse path     

Enhancing Fairness and Throughput of TCP in Heterogeneous Wireless Access Networks

Most of the recent research on TCP over heterogeneous wireless networks has concentrated on differentiating between packet drops caused by congestion and link errors, to avoid significant throughput degradations due to the TCP sending window being frequently shut down, in response to packet losses caused not by congestion but by transmission errors over wireless links. However, TCP also exhibits inherent unfairness toward connections with long round-trip times or traversing multiple congested routers. This problem is aggravated by the difference of bit-error rates between wired and wireless links in heterogeneous wireless networks. In this paper, we apply the TCP Bandwidth Allocation (TBA) algorithm, which we have proposed previously, to improve TCP fairness over heterogeneous wireless networks with combined wireless and wireline links. To inform the sender when congestion occurs, we propose to apply Wireless Explicit Congestion Notification (WECN). By controlling the TCP window behavior with TBA and WECN, congestion control and error-loss recovery are effectively separated. Further enhancement is also incorporated to smooth traffic bursts. Simulation results show that not only can the combined TBA and WECN mechanism improve TCP fairness, but it can maintain good throughput performance in the presence of wireless losses as well. A salient feature of TBA is that its main functions are implemented in the access node, thus simplifying the sender-side implementation.     

一种异构网络TCP拥塞控制算法

针对互联网中端对端带宽、时延和丢包率等的差异性日益加剧,导致TCP传输性能严重退化,该文提出一种链路自适应TCP拥塞控制算法(INVS)。INVS在拥塞避免阶段初期采用基于指数函数的凸窗口增长函数,以提高链路利用率;在窗口增长函数中引入了自适应增长因子实现窗口增长速率与链路状态相匹配;采用了自适应队列门限的丢包区分策略以提高无线环境下TCP的性能。性能分析和评估表明,INVS提高了TCP拥塞控制算法的吞吐量、公平性、链路利用率和RTT公平性。     

TFRC friendliness and the case of ECN

The TFRC protocol has been proposed as a TCP‐friendly protocol to transport streaming media over the Internet. However, its deployment is still questionable because it has not been compared to other important protocols, analysed in the presence of important mechanisms, such as the explicit congestion notification (ECN), and studied under more realistic network conditions. In this paper, we address these three aspects, including other congestion control protocols not considered before in the same investigation, such as TCP Tahoe, Reno, Newreno, Vegas, Sack, GAIMD, and the Binomial algorithms, the effect of using ECN in the friendliness of the protocols, and the fairness of the protocols under static and dynamic network conditions. We found that TFRC can be safely deployed in the Internet if competing with TCP Tahoe, New Reno and SACK since fairness is achieved under all scenarios considered. We also found that ECN actually helps in achieving better fairness. However, fairness problems arise when TFRC competes with TCP Reno, GAIMD, SQRT or IIAD in static or dynamic conditions, or both. We used normalized throughput, fairness index, and convergence time as the main performance metrics for comparison. Copyright © 2004 John Wiley & Sons, Ltd.     

TCP拥塞控制技术初探

首先探讨了网络拥塞出现的原因,然后分析了TCP拥塞控制的原理及四个TCP拥塞控制算法的性能,接着着重论述了TCP拥塞控制所面临的问题和对应算法的改进,最后提出了其进一步的研究方向。