High Bandwidth-delay Product Research Articles

A Cooperant Congestion Control Protocol in High Bandwidth-Delay Product Networks

PDF HTML阅读 XML下载 导出引用 引用提醒 高带宽延时网络中一种协同式拥塞控制协议 DOI: 作者: 作者单位: 作者简介: 通讯作者: 中图分类号: 基金项目: Supported by the National Natural Science Foundation of China under Grant Nos.90304010, 60673164 (国家自然科学基金); the Excellent Youth Foundation of Hu’nan Scientific Committee of China under Grant No.06JJ10009 (湖南省杰出青年基金); the Program for New Century Excellent A Cooperant Congestion Control Protocol in High Bandwidth-Delay Product Networks Author: Affiliation: Fund Project: 摘要 | 图/表 | 访问统计 | 参考文献 | 相似文献 | 引证文献 | 资源附件 | 文章评论 摘要:提出一种协同工作式的TCP(transmission control protocol)拥塞控制改进协议C3P(cooperant congestion control protocol),通过C3P源端检测RTT(round trip time)延时信息和路由器反馈的1 bit显式预测信息来判断网络拥塞状态,自适应地调节拥塞窗口.仿真实验表明,C3P协议能够有效地适应这种高带宽延时网络的传输特性,以保证网络获得更优的链路利用率、TCP友好 Abstract:This paper presents a Cooperant Congestion Control Protocol (C3P) that uses 1 bit routers' explicit feedback predicted information and delay signals to adjust the congestion windows appropriately. Simulation results show that C3P can efficiently improve the bandwidth utilization, TCP (transmission control protocol)-friendliness, RTT (round trip time) fairness and reduce the packet drop rate in High Bandwidth-Delay Product Networks. 参考文献 相似文献 引证文献

Quick Vegas: Improving Performance of TCP Vegas for High Bandwidth-Delay Product Networks

An important issue in designing a TCP congestion control algorithm is that it should allow the protocol to quickly adjust the end-to-end communication rate to the bandwidth on the bottleneck link. However, the TCP congestion control may function poorly in high bandwidth-delay product networks because of its slow response with large congestion windows. In this paper, we propose an enhanced version of TCP Vegas called Quick Vegas, in which we present an efficient congestion window control algorithm for a TCP source. Our algorithm improves the slow-start and congestion avoidance techniques of original Vegas. Simulation results show that Quick Vegas significantly improves the performance of connections as well as remaining fair when the bandwidth-delay product increases.

TCP-MT네트워크에서 다중 시간 간격을 이용한 혼잡제어 성능 개선에 관한 연구

Measurement of network traffic have shown that the self-similarity is a ubiquitous phenomenon spanning across diverse network environments. In previous work, we have explored the feasibility of exploiting the long-range correlation structure in a self-similar traffic for the congestion control. We have advanced the framework of the multiple time scale congestion control and showed its effectiveness at enhancing performance for the rate-based feedback control. Our contribution is threefold. First, we define a modular extension of the TCP-a function called with a simple interface-that applies to various flavours of the TCP-e.g., Tahoe, Reno, Vegas and show that it significantly improves performance. Second, we show that a multiple time scale TCP endows the underlying feedback control with proactivity by bridging the uncertainty gap associated with reactive controls which is exacerbated by the high delay-bandwidth product in broadband wide area networks. Third, we investigate the influence of the three traffic control dimensions-tracking ability, connection duration, and fairness-on performance.

Loss-resilient window-based congestion control

This paper addresses the problem of fair allocation of bandwidth resources on lossy channels in hybrid 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 build 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 feedback. A novel retransmission timer is proposed in order 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 analyze the performance of the new loss monitoring and recovery strategies, when used with two distinct explicit congestion control mechanisms. The first one relies on a coarse binary congestion notification from the routers. The second one, introduced in [D. Katabi, M. Handley, C. Rohrs, Internet congestion control for high bandwidth-delay product environments, ACM SIGCOMM (2002) 89-102], exploits accurate and finely-tuned router feedbacks to compute a precise congestion window adjustment. For both congestion control mechanisms, we observe that retransmissions triggered based on a precise monitoring of losses lead to 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 window-based congestion control, combined with appropriate error control strategies, can therefore provide a valid solution to reliable and controlled connections over lossy network infrastructures.

Satellite-3G Hybrid Networks: Impact of ACM and ARQ on TCP Performance

The adoption of satellite systems in providing broadband transmissions to mobile users such as trains, buses and vans is expected to be an interesting solution. The scenariowe considered refers to a hybrid network architecture, where a geostationary satellite forward link and a terrestrial 3G return link are used in order to exploit both the high bandwidth of a satellite channel and the lower propagation delay of a terrestrial path. The resulting round-trip delay is much shorter than that one experienced by using both the forward and return link via satellite. This is particularly appealing for overcoming the TCP efficiency degradation in high delay-bandwidth product and error prone channels. In this hybrid scenario, we used simulation results to compare the goodput of four of the most popular TCP variants, in the presence of a GOOD-BAD satellite channel, as the one experienced by mobile users. We applied an AdaptiveCoding and Modulation (ACM) technique as well, and studied its impact on TCP efficiency, when used both alone and in cooperation with an Automatic Repeat reQuest (ARQ) scheme of the Selective Repeat (SR) type with low persistency. Results obtained indicate that this hybrid architecture is advantageous for TCP transmissions in terms of average goodput, and that ACM is effective only if it is jointly used with ARQ schemes.

Exploring TCP Parallelisation for performance improvement in heterogeneous networks

TCP Parallelisation uses a set of parallel (modified or standard) TCP connections to transfer data for an application process. With standard TCP connections, TCP Parallelisation has been used to effectively utilise bandwidth for data intensive applications over high bandwidth-delay product (BDP) networks. On the other hand, it has been argued that a single TCP connection with proper modification can emulate and capture the robustness of TCP Parallelisation and thus can well replace TCP Parallelisation. A typical example is HSTCP, which can emulate the behaviour of a set of multiple standard TCP connections. However, through the analysis and the proposed use of TCP Parallelisation in this paper, we found that TCP Parallelisation has its unique merits, which cannot be emulated by a single-connection based approach. As a result, the single-connection based approach (such as HSTCP) may not be able to achieve the same effects as TCP Parallelisation, especially in heterogeneous networks and highly dynamic environments (e.g., high BDP networks combined with wireless access links). In this paper, we demonstrate some of these unique merits of TCP Parallelisation (namely, localisation of disturbances such as Fast Recovery, loss synchronisation between parallel connections and self- healing) over single-connection based approach in the heterogeneous environment.

On window control algorithm for TCP efficiency over wireless networks with high bandwidth-delay product

The existing congestion control algorithms such as TCP Reno are not suitable for efficient utilisation of networks with a high bandwidth-delay product because it takes a very long time to achieve the full link utilisation. Furthermore, it is difficult to guarantee the fairness among the TCP connections with different round-trip times. To overcome these problems, a new window control algorithm using the buffer state and variable gains is proposed. The simulation results show that our algorithm improves the performance of TCP in the wireless networks.

TCP-Ho: A Congestion Control Algorithm with Design and Performance Evaluation

A critical design issue of Transmission Control Protocol (TCP) is its congestion control that allows the protocol to adjust the end-to-end communication rate based on the detection of packet loss. However, TCP congestion control may function poorly during its slow start and congestion avoidance phases. This is because TCP sends bursts of packets with the fast window increase and the ACK-clock based transmission in slow start, and respond slowly with large congestion windows especially in high bandwidth-delay product (BDP) networks during congestion avoidance. In this article, we propose an improved version of TCP, TCP-Ho, that uses an efficient congestion window control algorithm for a TCP source. According to the estimated available bandwidth and measured round-trip times (RTTs), the proposed algorithm adjusts the congestion window size with a rate between exponential growth and linear growth intelligently. Our extensive simulation results show that TCP-Ho significantly improves the performance of connections as well as remaining fair and stable when the BDP increases. Furthermore, it is feasible to implement because only sending part needs to be modified.

An explicit router feedback framework for high bandwidth-delay product networks

Links with high bandwidth ranging from 1 to 10Gbps are increasingly in use worldwide. Congestion control with the positive use of router feedback that explicitly indicates network conditions is a promising way to address the performance issues of congestion control especially in such high-speed networks. In this paper, we propose SIRENS, a scalable, robust, and flexible fine-grained explicit router feedback framework. SIRENS is a per-hop and in-band notification scheme where each router captures a snapshot of the various kinds of downstream link status along the IP-level path from a sender to a receiver and notifies the receiver of the status. The receiver can find out the overall path status by assembling all the cumulative notifications that indicate the status in each of the single hops and by feedback can share this information with the sender. Such per-hop information is needed by end-hosts if we are to flexibly design novel congestion control mechanisms or to significantly improve the performance of conventional forms of congestion control. We show the feasibility of SIRENS in terms of router processing overhead through the development and evaluation of a network-processor-based high-performance network emulator. As a typical application of SIRENS, we show the configuration of TCP Limited Slow-Start. Through the implementation and evaluation, we have made sure that a sender can shift from the slow-start phase to the congestion avoidance phase without any packet drop, and can achieve more effective bandwidth utilization.

Free-Drop TCP

We introduce a new class of TCP congestion control algorithms that take a non-standard approach: instead of modifying AIMD parameters or exploiting traffic measurements, the new protocols modify the rule for deciding when to cut the congestion window. The class is defined by an additional window with a packet-count parameter w; the congestion window is reduced by half when a packet loss is detected, at time t say, if and only if there has been at least one dropped packet in the last w packet transmissions prior to time t . An algorithm in the class is called Free-Drop TCP , since dropped packets are "free" (they do not cause cuts in the window size) unless they are sufficiently bursty. We propose this new class as a means to achieve high utilizations in high bandwidth-delay product networks with small buffers. We analyze a fluid model which leads to explicit estimates of the average throughput for small loss probabilities. We then give the results of experiments, which show that, relative to TCP, a family of 'shifted' response functions of the form O (1/√ p - ε) can be obtained over a wide range of p by suitably varying w . Potential costs of these increases in throughput are also examined in terms of coefficents of variation and Jain's fairness measure. The costs range from negligible to moderate.

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.

XCP for shared-access multi-rate media

The eXplicit Control Protocol (XCP) was developed to overcome some of the limitations of TCP, such as low utilization in high bandwidth delay product networks, unstable throughput, large queue build-up, and limited fairness. XCP, however, requires that each queue controller in a path knows the exact capacity of its link. In shared access media, e.g. IEEE 802.11, knowing the actual capacity of the channel is a difficult task.In this paper we propose modifications to the XCP algorithm that enable the utilization of XCP even when the capacity of a link is unknown. These modifications are validated through simulation.We also present the results of a comparison between the performance of the modified XCP and TCP, where XCP controlled flows result more stable, fairness increases, and the network delay becomes lower. In addition, as the bandwidth delay product increases, XCP is able to maintain near-maximum utilization while TCP decreases utilization.

Comparison of FEC types with regard to the efficiency of TCP connections over AWGN satellite channels

Optimizing the end-to-end throughput of a TCP connection (goodput) over geostationary satellite links is a challenging research topic. This is because the high delay-bandwidth product, together with a non-negligible random loss of packets, is a condition that considerably differs from the environments TCP was originally designed for. As a result, TCP performance is significantly impaired by the channel bit error rate. The literature is full of suggestions for improving TCP goodput, most based on modifications of the protocol itself. We only investigated the application of different FEC (forward error correction) types for TCP goodput optimization, leaving the end-to-end protocol unaltered. Using a method midway between analysis and simulation to evaluate the goodput of TCP long-lived connections, we first studied the influence of packet loss rate, introduced by errors on the channel, on the TCP goodput. We showed that, in some cases, the packet loss rate does not need to be negligible with respect to that caused by congestion, as it is widely-held opinion. We then applied physical-level FEC techniques, such as convolutional encoding/Viterbi decoding, Reed Solomon, link-level erasure codes and their combinations, over a wide field of signal to noise conditions of the satellite channel. For each FEC type, we found the FEC rate that maximizes the TCP goodput, in each channel condition. We finally compared the results of all FECs used between them, and presented the case of multiple TCP connections sharing the same link as well

A dynamic approach to reduce preemption in book-ahead reservation in QoS-enabled networks

Increasing demand for multimedia and distributed applications in recent years has drawn renewed attention to book-ahead (BA) reservation. BA reservation has proved itself as an effective technique for time sensitive applications that require high amount of bandwidth with guaranteed QoS. However, BA reservation causes preemption of on-going instantaneous request (IR) calls when resource scarcity arises. Preemption disrupts service continuity for the preempted calls, specially in high speed large networks where rerouting is costly and often disabled because of high bandwidth-delay product and buffer capacity constraint. Recent studies on QoS definition has identified service continuity as one of the most important issues perceived by the users. Minimization of service disruption is crucial for effective implementation of BA reservation. Earlier models proposed to reduce IR call preemption use constant look-ahead time. This paper proposes a new model to calculate look-ahead time dynamically taking traffic parameters and network state into consideration. Simulation results show that the proposed model outperforms earlier model in terms of preemption probability, network utilization, and wasted throughput. The proposed model offers the flexibility of achieving a set level of preemption rate by appropriate selection of a single parameter value. Two techniques to estimate an appropriate value of this parameter are also presented.

One more bit is enough

Achieving efficient and fair bandwidth allocation while minimizing packet loss in high bandwidth-delay product networks has long been a daunting challenge. Existing end-to-end congestion control ( eg TCP) and traditional congestion notification schemes ( eg TCP+AQM/ECN) have significant limitations in achieving this goal. While the recently proposed XCP protocol addresses this challenge, XCP requires multiple bits to encode the congestion-related information exchanged between routers and end-hosts. Unfortunately, there is no space in the IP header for these bits, and solving this problem involves a non-trivial and time-consuming standardization process.In this paper, we design and implement a simple, low-complexity protocol, called Variable-structure congestion Control Protocol (VCP), that leverages only the existing two ECN bits for network congestion feedback, and yet achieves comparable performance to XCP, ie high utilization, low persistent queue length, negligible packet loss rate, and reasonable fairness. On the downside, VCP converges significantly slower to a fair allocation than XCP. We evaluate the performance of VCP using extensive ns2 simulations over a wide range of network scenarios. To gain insight into the behavior of VCP, we analyze a simple fluid model, and prove a global stability result for the case of a single bottleneck link shared by flows with identical round-trip times.

A rate control scheme for adaptive real-time applications in IP networks with lossy links and long round trip times

Currently there is no control for real-time traffic sources in IP networks. This is a serious problem because real-time traffic can not only congest the network but can also cause unfairness and starvation of TCP traffic. However, it is not possible to apply current solutions for Internet to the networks with high bandwidth-delay products and high bit error rates. The channel errors may result in inaccurate congestion control decisions and unnecessary rate throttles leading to severe performance degradation. This problem is amplified in the links with high bandwidth-delay products, since the link is inefficiently utilized for a very long time until the unnecessary rate throttle is recovered. In this paper, a new Rate Control Scheme, RCS, is introduced for real-time interactive applications in networks with high bandwidth-delay products and high bit error rates. RCS is based on the concept of using dummy packets to probe the availability of network resources. Dummy packets are treated as low priority packets and consequently they do not affect the throughput of actual data traffic. Therefore, RCS requires all the routers in the connection path to support some priority policy. A new algorithm is also proposed to improve the robustness of the RCS to temporal signal loss conditions. The delay-bound considerations for real-time traffic sources using RCS rate control scheme are also investigated. Simulation experiments show that in environments with high bandwidth-delay products and high bit error rates, RCS achieves high throughput performance without penalizing TCP connections.

Analysis of MIMD congestion control algorithm for high speed networks

Proposals to improve the performance of TCP in high speed networks have been recently put forward. Examples of such proposals include High Speed TCP, Scalable TCP, and FAST. In contrast to the additive increase multiplicative decrease algorithm used in the standard TCP, Scalable TCP uses a multiplicative increase multiplicative decrease (MIMD) algorithm for the window size evolution. In this paper, we present a mathematical analysis of the MIMD congestion control algorithm in the presence of random losses. Random losses are typical to wireless networks but can also be used to model losses in wireline networks with a high bandwidth-delay product. Our approach is based on showing that the logarithm of the window size evolution has the same behaviour as the workload process in a standard G/G/1 queue. The Laplace–Stieltjes transform of the equivalent queue is then shown to directly provide the throughput of the congestion control algorithm and the higher moments of the window size. Using ns-2 simulations, we validate our findings using Scalable TCP.

Teraflows over Gigabit WANs with UDT

The TCP transport protocol is currently inefficient for high speed data transfers over long distance networks with high bandwidth delay products (BDP). The challenge is to develop a protocol which is fast over networks with high bandwidth delay products, fair to other high volume data streams, and friendly to TCP-based flows. We describe here a UDP-based application level transport protocol named UDT (UDP-based Data Transfer) with these properties and which is designed to support distributed data-intensive computing applications. UDT can utilize high bandwidth efficiently over wide area networks with high bandwidth delay products. Unlike TCP, UDT is fair to flows independently of their round trip times (RTT). In addition, UDT is friendly to concurrent TCP flows, which means it can be deployed not only on experimental research networks but also on production networks. To ensure these properties, UDT employs a novel congestion control approach that combines rate-based and window-based control mechanisms. In this paper, we describe the congestion control algorithms used by UDT and provide some experimental results demonstrating that UDT is fast, fair, and friendly.

Experimental studies of data transport and data access of earth-science data over networks with high bandwidth delay products

Although the amount of earth-science data is growing rapidly, as is the availability of high-performance networks, our ability to access large remote earth-science data sets is still very limited. This is particularly true of networks with high bandwidth-delay products (BDP), such as those between the US and Europe. Recently, several network protocols have emerged that improve the situation and hold the promise of being much more effective than striped TCP (wherein multiple, parallel connections are used to increase data throughput). In this paper, we report on experimental studies using one of these new protocols called UDT, and compare UDT to other approaches. In addition, we consider the effectiveness of these new protocols when reading and writing data to and from disk over high BDP networks. We also consider the problem of accessing remote data by attribute over these same networks. We show that with the appropriate protocol, the speed of accessing data across the Atlantic can be improved significantly. We note that the UDT protocol used here can be deployed as an application library for earth-science applications and neither requires upgrades to existing network infrastructure, such as routers, nor to the Linux kernels on the servers involved.

Performance Evaluation of TCP over Optical Channels and Heterogeneous Networks

Next generation optical networks will soon provide users the capability to request and obtain end-to-end all optical 10 Gbps channels on demand. Individual users will use these channels to exchange large amounts of data and support applications for scientific collaborative work. These new applications, which expect steady transfer rates in the order of Gbps, will very likely use either TCP or a new transport layer protocol as the end-to-end communication protocol. In this paper, we investigate the performance of TCP and newer TCP versions over High Bandwidth Delay Product Channels (HBDPC), such as the on demand optical channels described above. In addition, we investigate the performance of these new TCP versions over wireless networks and according to old issues such as fairness. This is particularly important to make adoption decisions. Using simulations, we show that (1) the window-based mechanism of current TCP implementations is not suitable to achieve high link utilization and (2) congestion control mechanisms, such as the one utilized by TCP Vegas and Westwood are more appropriate and provide better performance. We also show that new TCP proposals, although perform better than current TCP versions, they still perform worse than TCP Vegas. In addition, we found that even though these newer versions improve TCP's performance over their original counterparts in HBDPC, they still have performance problems in wireless networks and present worse fairness problems than their old counterparts. We conclude that all these versions are still based on TCP's AIMD strategy or similar and therefore continue to be fairly blind in the way they increase and decrease their transmission rates. TCP will not be able to utilize the foreseen optical infrastructure adequately and support future applications if not redesigned to scale.