High-speed TCP Variants Research Articles

위성 통신 시스템에서 TCP연결 분할 기반 PEP의 성능 평가

A satellite communication system is one of viable solutions for Internet applications running in wide areas. However, the performance of TCP can be seriously degraded in the satellite networks due to long round-trip time (RTT) and high bit error rate (BER) over satellite links. Therefore, a performance enhancing proxy(PEP) based TCP splitting connection scheme is used in the satellite link to improve the TCP performance. In this paper, we implement PEP testbed and conduct experiment to evaluate the performance of TCP splitting connection by comparing with high-speed TCP solutions in various environments. In our experimental environment, we consider multiple connections, high packet loss, and limited bandwidth. The experiment results show that PEP improves the TCP throughput than high-speed TCP variants in various environments. However, there is no improvement of the TCP throughput with the limited bandwidth because there is packet loss caused by both the congestion and the channel error.

Impact of Queue Management Schemes and TCP Variants on the Performance of 10Gbps High Speed Networks: An Experimental Study

Queue management schemes at routers and congestion avoidance schemes at end points cooperate to provide good congestion solutions in computer networks. While queue management schemes are still being developed, research on congestion avoidance has come a long way to serve the bandwidth requirement of the networks (e.g. high speed networks, data centers, etc.) at the order of 10Gbps. Because of considerable lack of the evaluation research work, there is no consensus on the choice of the queue management algorithms over these networks. To the best of our knowledge, this is the first work that presents the experimental study of the effect of various queue management schemes on high speed TCP variants in a 10Gbps network environment. Evaluations of queue management schemes such as Drop-tail, RED, CHOKe, and SFB are presented with popular high speed TCP variants such as RENO, HSTCP, CUBIC, and VEGAS over CRON, a 10Gbps high speed network testbed. Performance results are presented for several important metrics of interest such as link utilization, intro-protocol fairness, RTT fairness, delay and computational complexity. We argue the importance of explicit consideration of the basic tradeoffs between TCP variants and router parameters that network designers must face when building high speed networks.

A study of fairness among heterogeneous TCP variants over 10 Gbps high-speed optical networks

Several high-speed TCP variants are adopted by end users and therefore, heterogeneous congestion control has become the characteristic of newly emerging high-speed optical networks. In contrast to homogeneous TCP flows, fairness among heterogeneous TCP flows now depends on router parameters such as queue management scheme and buffer size. To the best of our knowledge, this is the first study of fairness performance among heterogeneous TCP variants over a 10Gbps high-speed optical network environment. Our evaluation scenarios for heterogeneous TCP flows consist of TCP variants with substantial presence in current Internet; therefore, TCP-SACK, CUBIC TCP (CUBIC) and HighSpeed TCP (HSTCP) compete for bottleneck bandwidth. Experimental results for fairness are presented for different queue management schemes, such as Drop-tail, Random Early Detection (RED), CHOose and Keep for responsive flows CHOose and Kill for unresponsive flows (CHOKe), and Approximate Fair Dropping (AFD), with varying degree of buffer sizes. We observe heterogeneous TCP flows induce more unfairness than homogeneous ones. Active queue management (AQM) schemes, RED, CHOKe, and AFD, improve fairness for large buffer sizes as compared to Drop-tail, whereas AQM schemes lose the fairness advantage for small buffer sizes. This study provides preliminary results for fairness challenges to deploy all-optical routers with limited buffer size.

A New Survey on Improving TCP Performances over Geostationary Satellite Link

The idea of Internet everywhere makes the assumption that an Internet access should be available even in remote areas without network infrastructure. In this case satellite access represents an attractive solution. Nevertheless, experience shows that over satellite links, TCP is limited in terms of data speeds. Many enhancements and solutions, based for instance on tuning TCP parameters or TCP spoofing, have been proposed to avoid the underutilization of satellite link capacity. These topics have been often addressed, but considering recent high speed TCP variants, the evolution of end users habits, and recently proposed satellite link access scheme, a new study is necessary in order to reconsider some preconceptions and previous recommendations in such a context. This paper proposes an overview of TCP variants and a survey of commonly proposed solutions for TCP over satellite. Then a methodology for TCP performance assessment over satellite links is exposed. The approach is mainly based on a satellite link emulation platform and some tools developed at the ENAC. We assess the gain offered by a split TCP connection with PEPSal comparatively to end to end TCP connections based on NewReno and a recent widely deployed TCP version (TCP Cubic) on an emulated satellite link. Unlike existing studies, we compare PEP advantages with most recent TCP versions and propose as an extension, to assess PEP gain considering the number of simultaneous TCP connections. Finally, the results provided allow us to make some original recommendations toward TCP deployment over satellite links.

Comparative Study of High-Speed TCP Variants in Multi-Hop Wireless Networks

Comparative Study of High-Speed TCP Variants in Multi-Hop Wireless Networks

Hybrid congestion control for high-speed networks

Nowadays, more and more applications require fast transfer of massive data over networks, and the emergence of high-speed networks provides an ideal solution to this challenge. Due to the limitations of the conservative congestion control algorithm, the standard TCP is no longer appropriate for high-speed networks to efficiently utilize the bandwidth resources. Therefore, several high-speed TCP variants have been suggested to conquer the problem. However, although these protocols perform successfully to improve the bandwidth utilization, they still have the weakness on the performance such as RTT-fairness, TCP-friendliness, etc. In this paper, we propose HCC TCP, a hybrid congestion control algorithm using the synergy of delay-based and loss-based approach for the adaptation to high speed and long distance network environment. The algorithm uses queuing delay as the primary congestion indicator, and adjusts the window to stabilize around the size which can achieve the full utilization of available bandwidth. On the other hand, it uses packet loss as the second congestion indicator, and a loss-based congestion control strategy is utilized to maintain high bandwidth utilization in the cases that the delay-based strategy performs inefficiently in the networks. The two approaches in the algorithm are dynamically transferred into each other according to the network status. We finally perform simulations to verify the properties of the proposed HCC TCP. The simulation results demonstrate HCC TCP satisfies the requirements for an ideal TCP variant in high-speed networks, and achieves efficient performance on throughput, fairness, TCP-friendliness, robustness, etc.

A loss-event driven scalable fluid simulation method for high-speed networks

Increase of size and bandwidth of computer network posed a research challenge to evaluate proposed TCP/IP protocol and corresponding queuing policies in this scenario. Simulation provides an easier and cheaper method to evaluate TCP proposals and queuing disciplines as compared to experiment with real hardware. In this paper, problem associated with scalability of current simulation method for high-speed network case is discussed. Hence, we present a scalable time-adaptive numerical simulation driven by loss events to represent dynamics of high-speed networks using fluid-based models. The new method uses a loss event to dynamically adjust the size of a time step for a numerical solver which solves a system of differential equations representing dynamics of protocols and nodes’ behaviors. A numerical analysis of the proposed protocol is discussed. A simple simulation of high-speed TCP variants is presented using our method. The simulation results and analysis show that the time-adaptive method reduces computational time while achieving the same accuracy compared to that of a fixed step-size method.

Perspectives on router buffer sizing

The past few years have witnessed a lot of debate on how large Internet router buffers should be. The widely believed rule-of-thumb used by router manufacturers today mandates a buffer size equal to the delay-bandwidth product. This rule was first challenged by researchers in 2004 who argued that if there are a large number of long-lived TCP connections flowing through a router, then the buffer size needed is equal to the delay-bandwidth product divided by the square root of the number of long-lived TCP flows. The publication of this result has since reinvigorated interest in the buffer sizing problem with numerous other papers exploring this topic in further detail - ranging from papers questioning the applicability of this result to proposing alternate schemes to developing new congestion control algorithms, etc. This paper provides a synopsis of the recently proposed buffer sizing strategies and broadly classifies them according to their desired objective: link utilisation, and per-flow performance. We discuss the pros and cons of these different approaches. These prior works study buffer sizing purely in the context of TCP. Subsequently, we present arguments that take into account both real-time and TCP traffic. We also report on the performance studies of various high-speed TCP variants and experimental results for networks with limited buffers. We conclude this paper by outlining some interesting avenues for further research.

Packet reordering in high-speed networks and its impact on high-speed TCP variants

Several recent Internet measurement studies show that the higher the packet sending rate, the higher the packet-reordering probability. This implies that recently proposed high-speed TCP variants are more likely to experience packet reordering than regular TCP in high-speed networks, since they are designed to achieve much higher throughput than regular TCP in these networks. In this paper, we first study the characteristics of packet reordering in high speed networks. Second, we verify the impact of packet reordering on high speed TCP variants and evaluate the effectiveness of the existing reordering-tolerant TCP enhancements using simulations. Our simulation results demonstrate that high-speed TCP variants perform poorly in the presence of packet reordering, and existing reordering-tolerant algorithms can significantly improve the performance of high-speed TCP variants.

Can high-speed networks survive with DropTail queues management?

It has been observed that TCP connections that go through multiple congested links (MCL) have a smaller transmission rate than the other connections. Such TCP behavior is a result of two components (i) the cumulative packet losses that a flow experiences at each router along its path; (ii) the longer round trip times (RTTs) suffered by such flows due to non-negligible queueing delays at congested routers. This double ''bias'' against connections with MCLs has been shown to approximate the so-called minimum potential delay fairness principle in the current Internet. Despite the recent proliferation of new congestion control proposals for TCP in high-speed networks, it is still unclear what kind of fairness principle could be achieved with such newly proposed congestion control protocols in high-speed networks with large-delays. Studies already show that some high-speed TCP variants may cause surprisingly severe RTT unfairness in high-speed networks with DropTail routers. This paper studies the problem of unfairness in high-speed networks with some well-known high-speed TCP variants in presence of multiple congested links and highlights the severity of such unfairness when DropTail queue management is adopted. Through a simple synchronized loss model analysis, we show how synchronized losses with DropTail in high-speed networks could lead to severe RTT unfairness and drop probability (DP) unfairness; while random marking AQM schemes, which break the packet loss synchrony mitigate such unfairness dramatically by ensuring that the packet loss probability of a flow is the sum of the loss probabilities on the congested routers it crosses. Extensive simulations are carried out and the results support our findings.

Experimental Results of Implementing High-Speed and Parallel TCP Variants for Long Fat Networks

As computer hardware components are achieving greater speeds, network link bandwidths are becoming wider. A number of enhancements to TCP have been developed in order to fully exploit these improvements in network infrastructures, including TCP window scale option, SACK option, and Highspeed TCP (HSTCP) modifications. However, even with these enhancements, TCP cannot provide satisfactory performance in high-speed long-delay networks. As a means addressing this problem, gentle Highspeed TCP (gHSTCP) has been proposed in [1]. However, its effectiveness has only been demonstrated in simulation experiments. In the present paper, a refined gHSTCP algorithm is proposed for application to real networks. The performance of the refined gHSTCP algorithm is then assessed experimentally. The refined gHSTCP algorithm is based on the original algorithm, which uses two modes (Reno mode and HSTCP mode) in the congestion avoidance phase and switches modes based on RTT increasing trends. The refined gHSTCP algorithm compares two RTT thresholds and judges which mode will be used. The performance of gHSTCP is compared with TCP Reno/HSTCP and parallel TCP mechanisms. The experimental results demonstrate that gHSTCP can provide a better tradeoff in terms of utilization and fairness against co-existing traditional TCP Reno connections, whereas HSTCP and parallel TCP suffer from the trade-off problem.