TCP Versions Research Articles

Rethinking Fast and Friendly Transport in Data Center Networks

The sustainable growth of bandwidth has been an inevitable tendency in current Data Center Networks (DCN). However, the dramatic expansion of link capacity offers a remarkable challenge to the transport layer protocols of DCN, i.e., how to converge fast and enable data flow to utilize the high bandwidth effectively. Meanwhile, the new protocol should be compatible to the traditional TCP because the applications with old TCP versions are still widely deployed. Therefore, it is important to achieve a trade-off between the aggressiveness and TCP-friendliness in protocol design. In this article, we first empirically investigate why the existing typical data center TCP variants naturally fail to guarantee both fast convergence and TCP friendliness. Then, we design a new transport protocol for DCN, namely Fast and Friendly Converging (FFC), which makes independent decisions and self-adjustment through retrieving the two-dimensional congestion notification from both RTT and ECN. We further present a mathematic model to analyze its competing behavior and converging process. The results from simulation experiments and real implementation show that FFC can achieve fast convergence, thus benefiting the flow completion time. Moreover, when coexisting with the traditional TCP, FFC also presents a moderate behavior, while introducing trivial deployment overhead only at the end-hosts.

Relationship Between Critical Pedagogy and Reflective Thinking with L2 Teachers’ Pedagogical Success

This study first sought to develop a survey (Teacher Critical Pedagogy, TCP) to evaluate L2 teachers’ degree of critical pedagogy. Second, it investigated the relationship between L2 teachers’ degree of critical pedagogy and reflective thinking with their pedagogical success. Third, it explored the extent to which teachers’ critical pedagogy and reflective thinking contributed to their pedagogical success. To these ends, factor analysis with the data collected from a sample of 397 Iranian EFL teachers was used through AMOS software to confirm the validity of the critical pedagogy questionnaire. Then, validated version of TCP, Kember, Leung, Jones, and Loke’s (2000) Reflective Thinking for Teachers, and Moafian and Pishghadam’s (2009) Characteristics of Successful Teachers questionnaires were administered to a sample of 53 EFL teachers and 503 EFL students in language institutes in Isfahan, Iran, to assess the teachers’ critical pedagogy and reflective thinking as well as their pedagogical success as evaluated by their students. Correlational and multiple regression analysis revealed a significant and positive relationship between the teachers’ critical pedagogy and reflective thinking with their pedagogical success. Moreover, critical pedagogy and reflective thinking predicted the teachers’ pedagogical success to a considerable extent. Findings provide implications for L2 teachers and teacher educators.

Discrete-Time Sliding-Mode Controllers for MPTCP Networks

TCP provides the principal framework for the Internet traffic control. Following the protocol development, however, the communicating parties have not benefitted from the potential of multiple interfaces, now commonly installed at the end-point devices. In order to grasp new opportunities, a multipath version of TCP - multipath TCP (MPTCP) - has recently been introduced. It permits the simultaneous use of a few, diverse interfaces for enhancing the user experience. This article provides a discrete-time model of the data exchange process in the MPTCP framework and new flow control algorithms. The algorithms, applicable to the paths with nonhomogenous properties with respect to delays, loss rate, etc., are designed following the rules of discrete sliding-mode control (DSMC). As a result, direct implementation in discrete networking context and robust transmission system performance are guaranteed in uncertain operating conditions. It is formally demonstrated that a finite data queue length in the end-point buffers and a feasible (non-negative and bounded) input signal are ensured despite input and output perturbations affecting the control process. It is also shown that the classical DSMC design approach results in decreased performance for unanticipated surge in the intensity of network-related perturbations. A second, improved algorithm, with a well-chosen sliding variable, eliminates this drawback.

RAPID: Avoiding TCP Incast Throughput Collapse in Public Clouds With Intelligent Packet Discarding

Many applications in public clouds require a high fan-in, many-to-one type of data communication (known as TCP incast) in modern Data Center Networks (DCNs). Such communication could cause severe incast congestion in switches and result in TCP throughput collapse, substantially degrading the application performance. The root cause of throughput collapse is the Retransmission Timeouts (RTO) due to packet losses in congested switches. Tenants in public clouds can opt to use a variety of TCP versions. However, the existing solutions rely on modifications of TCP protocols and specific techniques from switches, and thus these existing solutions are not always feasible for public clouds. In this paper, we are inspired by the emerging virtualization and network softwarization technologies to develop a novel scheme called Retransmission timeout Avoidance by Packet Intelligent Discarding (RAPID) using software switches. RAPID considers the number of packets of each incast flow, buffered in the switch to selectively discard some packets, and ensures that the Fast Retransmission/Fast Recovery rather than RTO is invoked at the sender(s) in response to packet loss. Thus, the long idle period of a timeout and the throughput drop are avoided. We prove that, given a predetermined minimum switch buffer space, dedicated to the incast application, RAPID can prevent RTO in all the incast senders. We also present a low-complexity heuristic version of RAPID named RAPID-ED, which combines the principles of RAPID and early detection and is extremely easy to implement on today’s software switches. We evaluate the two proposed schemes in a data center network testbed built on NS-3 simulator. The simulation results confirm the theoretical expectation, and show that the RAPID and RAPID-ED perform very well to prevent RTO of TCP incast flows and hence the throughput collapse. Compared with other incast solutions, RAPID and RAPID-ED do not modify TCP protocols and therefore are more suitable in public clouds.

An Analysis of IoT Congestion Control Policies

The Internet of Things (IoT) is common platform provide connectivity between various type of devices are having communication capability. Main objective of IoT is to deploy effective and high-quality smart services which offer different type services. Therefore, IoT need an advanced protocol stack which provides inter-communications between different types of devices have differet service requirements. IoT uses TCP and UDP as transport layer protocols to achieve the quality of service (QoS) required by different IoT devices. The XMPP (Extensible Messaging and Presence Protocol), MQTT (MQ Telemetry Transport) and RESTful HTTP are IoT application protocols which use TCP as a transport layer protocol. The transport layer protocol offers service point addressing, flow control, congestion control and segmentation, reassembling. As a number of different types of device increases over the Internet, network congestion also increases. In current internet scenario, TCP is responsible for congestion control over the Internet. However, current available TCP versions are not enough capable to handle different type of devices to fulfil the need of IoT. Therefore, this paper provides a survey of various congestion control algorithms used at transport layer. This paper includes available congestion control algorithms, their advantages, disadvantages and existing problems with TCP in IoT domains.

PERFORMANCE COMPARISON OF THE DIFFERENT STREAMS IN A TCP BOTTLENECK LINK IN THE PRESENCE OF BACKGROUND TRAFFIC IN A DATA CENTER

The purpose of this work is the evaluation of the performance of TCP streams when the link that causes a bottleneck is also shared by the background traffic. Standard TCP is the most used protocol which sustains the majority of Internet traffic. Nevertheless, TCP manifests some problems when using almost all of the available bandwidth. Based on these problems, there are recognized different versions of TCP. The aim of this paper is to test the conflict between TCP flows bandwidth allocation. We have implemented a network which includes FTP traffic and background traffic. The TCP flows are simulated so that they begin transmitting at different times. Background traffic is added besides the TCP flows so that we approach a real network model. Besides studying how the TCP flows compete with each other, we will evaluate if the background traffic has an impact on the behavior of TCP flows and if it influences how the available bandwidth is shared equally among flows. We compare standard TCP Tahoe and TCP Reno, which do not differ much between each-other, but use different algorithms so through simulation we will evaluate the changes in the bandwidth they use, even though we expect the TCP flows to be more aggressive in getting bandwidth from other TCP flows.

Mitigating incast-TCP congestion in data centers with SDN

In data center networks (DCNs), the presence of long-lived TCP flows tends to bloat the switch buffers. As a consequence, short-lived TCP-incast traffic suffers repeated losses that often lead to loss recovery via timeout. Because the minimum retransmission timeout (minRTO) in most TCP implementations is fixed to around 200 ms, interactive applications that often generate short-lived incast traffic tend to suffer unnecessarily long delays waiting for the timeout to elapse. The best and most direct solution to such problem would be to customize the minRTO to match DCNs delays; however, this is not always possible; in particular in public data centers where multiple tenants, with various versions of TCP, co-exist. In this paper, we propose to achieve the same result by using techniques and technologies that are already available in most commodity switches and data centers and that do not interfere with the tenant’s virtual machines or TCP protocol. In this approach, we rely on the programmable nature of SDN switches and design a SDN-based incast congestion control (SICC) framework, that uses a SDN network application in the controller and a shim-layer in the host hypervisor, to mitigate incast congestion. We demonstrate the performance gains of the proposed scheme via real deployment in a small-scale testbed as well as ns2 simulation experiments in networks of various sizes and settings.

Steady-state performance evaluation of Linux TCPs versus TCP wave over leaky satellite links

Internet access is becoming even more heterogeneous, including different wireless backhauling links, with Ka-band satellites as a possible alternative. Since communication path is unknown a priori, adoption of PEP solutions to optimize TCP performance over satellite is discouraged to allow dynamic network reconfigurations. To opposite, an end-to-end TCP performance evaluation on such a challenging scenario, with possible large latency and transmission losses, is herein considered of paramount importance. Several TCP variants exist to tackle different aspects of communication networks. In Linux, the different TCP congestion control schemes differ from the theoretical formulations and RFC specifications, introducing a varying set of optimizations and options. This aspect makes difficult to identify a standard/reference TCP version for the proposed scenario, while testing with the real protocol stacks is deemed necessary to obtain consistent results. In addition, an innovative end-to-end TCP, namely TCP Wave, is introduced to replace the traditional window-based transmission with a burst-based strategy, representing a valid alternative to Linux TCP. To offer a fair, realistic and comprehensive evaluation, we configured a simulation setup where different Linux TCPs can be run within ns-3 network simulator and compared with TCP Wave.

Effective Internet Traffic Management by Reducing Congestion in TCP Cubic through Proactive Approach

The future of modern Internet is adapting Cloud Computing, IoT and 3D technology over the Internet. So, Internet needs to exchange a huge amount of information in the form of packets of requests, responses and controlling data. An Internet connection always has finite throughput and can only support certain number of packets at a time. Transferring too many packets on Internet connection can produce heavy Internet Traffic which may cause bottleneck or network congestion. Meanwhile, the operations between the communication devices also effect and slow down. Effective congestion control is one of the most critical issues in the efficiency of Internet Traffic Management. To make it sustainable and efficient, a systematic approach to planning, testing and maintaining its behavioral mechanism is desirable to avoid network congestion and keep the data flowing smoothly. In this area of traffic management, the research will be conducted on different TCP variants and proposed a model of congestion avoidance. The new Congestion Avoidance algorithmic model will enhance the performance and utilization of the network, by making TCP more flexible and efficient protocol with a new proactive approach. The results of analysis will help the industrial developers and researchers to implement the proposed solution in Linux Kernel for more stable version of TCP.

Performance Analysis of TCP NewReno over a Cellular Last-Mile: Buffer and Channel Losses

TCP NewReno is one of the most widely deployed TCP versions in today's Internet. However, a full understanding of the complex inter-dependencies between the losses due to wireless channel errors and those due to buffer overflows, and their (joint) impact on TCP NewReno's congestion control algorithm in wireless and wired-cum-wireless networks is still lacking. In this paper, we develop a comprehensive analytical model for, and study the performance of, TCP NewReno with a cellular last-mile access, taking into account both types of losses. We assume a frame-level Markovian loss model, and build a model that features the system's basic controllable parameters (such as the number of retransmissions and the buffer size), so as to study how they (jointly) affect the TCP-level throughput. We model certain finer aspects, e.g., correlations in wireless and buffer losses and their cross-correlation. We provide a summary of numerical results highlighting several non-trivial findings. In particular, we demonstrate that there exist optimal (i.e., TCP throughput maximizing) pairs of the number of retransmissions and the buffer size.

Research on TCP Congestion Control Mechanisms

With the rapid development of information society, the number of information data transmitting in the Internet is growing quickly, so congestion control becomes very important and it is a hot research field on the Internet now. In this paper, the sliding window, reasons and TCP algorithms of congestion control such as slow start, congestion avoidance, fast retransmit and fast recovery are discussed. Different TCP versions about congestion control are studied and analyzed. As a result, improvements on the congestion control algorithm and the research is still an important topic in the study of Internet congestion control.

Cross-layer TCP performance analysis in IEEE 802.11 vehicular environments

In this paper we provide a performance analysis of TCP in IEEE 802.11 vehicular environments for different well-known TCP versions, such as Tahoe, Reno, New Reno, Vegas, and Sack. The parameters of interest from the TCP side are the number of Duplicate Acknowledgements - DupAck, and the number of Delayed Acknowledgements - DelAck, while on the wireless network side the analyzed parameter is the interface queue - IFQ. We have made the analysis for the worst-case distance scenario for single-hop and worst-case multihop vehicular environments. The results show that the number of wireless hops in vehicular environments significantly reduces the TCP throughput. The best average performances considering all scenarios were obtained for TCP Vegas. However, the results show that the interface queue at wireless nodes should be at least five packets or more. On the other side, due to shorter distances in the vehicular wireless network, results show possible flexibility of using different values for the DupAck without degradation of the TCP throughput. On the other side, the introduction of the DelAck parameter provides enhancement in the average TCP throughput for all TCP versions.

Transfer Reliability and Congestion Control Strategies in Opportunistic Networks: A Survey

Opportunistic networks are a class of mobile ad hoc networks (MANETs) where contacts between mobile nodes occur unpredictably and where a complete end-to-end path between source and destination rarely exists at one time. Two important functions, traditionally provided by the transport layer, are ensuring the reliability of data transmission between source and destination, and ensuring that the network does not become congested with traffic. However, modified versions of TCP that have been proposed to support these functions in MANETs are ineffective in opportunistic networks. In addition, opportunistic networks require different approaches to those adopted in the more common intermittently connected networks, e.g. deep space networks. In this article we capture the state of the art of proposals for transfer reliability and storage congestion control strategies in opportunistic networks. We discuss potential mechanisms for transfer reliability service, i.e. hop-by-hop custody transfer and end-to-end return receipt. We also identify the requirements for storage congestion control and categorise these issues based on the number of message copies distributed in the networks. For single-copy forwarding, storage congestion management and congestion avoidance mechanism are discussed. For multiple-copy forwarding, the principal storage congestion control mechanisms are replication management and drop policy. Finally, we identify open research issues in the field where future research could usefully be focused.

Got loss? Get zOVN!

Datacenter networking is currently dominated by two major trends. One aims toward lossless, flat layer-2 fabrics based on Converged Enhanced Ethernet or InfiniBand, with benefits in efficiency and performance. The other targets flexibility based on Software Defined Networking, which enables Overlay Virtual Networking. Although clearly complementary, these trends also exhibit some conflicts: In contrast to physical fabrics, which avoid packet drops by means of flow control, practically all current virtual networks are lossy. We quantify these losses for several common combinations of hypervisors and virtual switches, and show their detrimental effect on application performance. Moreover, we propose a zero-loss Overlay Virtual Network (zOVN) designed to reduce the query and flow completion time of latency-sensitive datacenter applications. We describe its architecture and detail the design of its key component, the zVALE lossless virtual switch. As proof of concept, we implemented a zOVN prototype and benchmark it with Partition-Aggregate in two testbeds, achieving an up to 15-fold reduction of the mean completion time with three widespread TCP versions. For larger-scale validation and deeper introspection into zOVN, we developed an OMNeT++ model for accurate cross-layer simulations of a virtualized datacenter, which confirm the validity of our results.

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.

Transport layer enhancements on a satellite‐based mobile broadband system for high‐speed trains

SUMMARYThe great diffusion of wireless portable and handheld devices, along with the growing availability of applications on the Internet, led to a significant increase in the demand of access. Users require to continue business or private activities when moving, with performance at least comparable with that experienced over fixed networks. A very interesting scenario encompasses broadband and uninterrupted access over high‐speed trains. Developments of new telecommunication standards as well as innovative network architectures give the potential to fully match the aforementioned requirements. Nevertheless, there is a significant number of issues, especially related to system integration, mobility management, and propagation channel conditions, that need a careful analysis with the aim to support the design of real systems. This paper deals with this challenging scenario proposing a satellite‐based architecture in which DVB‐S2 and DVB‐RCS standards, MIPv6 at layer 3 (for handover management), and an enhanced TCP version at layer 4, named TCP Noordwijk, allow to achieve good performance. An NS‐2 enhanced simulator has supported the whole analysis. Results show the effectiveness of the proposed scheme against different modulation and coding configurations and compared with a selected set of TCP‐enhanced versions. Copyright © 2012 John Wiley & Sons, Ltd.

A Multipath Cubic TCP Congestion Control with Multipath Fast Recovery over High Bandwidth-Delay Product Networks

Cubic TCP, one of transport protocols designed for high bandwidth-delay product (BDP) networks, has successfully been deployed in the Internet. Multi-homed computers with multiple interfaces to access the Internet via high speed links will become more popular. In this work, we introduce an extended version of Cubic TCP for multiple paths, called MPCubic. The extension process is approached from an analysis model of Cubic by using coordinated congestion control between paths. MPCubic can spread its traffic across paths in load-balancing manner, while preserving fair sharing with regular TCP, Cubic, and MPTCP at common bottlenecks. Moreover, to improve resilience to link failure, we propose a multipath fast recovery algorithm. The algorithm can significantly reduce the recovery time of data rate after restoration of failed links. These techniques can be useful for resilient high-bandwidth applications (for example, tele-health conference) in disaster-affected areas. Our simulation results show that MPCubic can achieve stability, throughput improvement, fairness, load-balancing, and quick data rate recovery from link failure under a variety of network conditions. key words: multipath congestion control, Cubic TCP for multiple paths, load-balancing, resource pooling, fairness, resilience, disaster

Improve Performance of TCP New Reno Over Mobile Ad-Hoc Network Using ABRA

In a mobile ad hoc network, temporary link failures and route changes occur frequently. With the assumption that all packet losses are due to congestion, TCP performs poorly in such an environment. There are many versions of TCP which modified time to time as per need. In this paper modifications introduced on TCP New Reno over mobile ad-hoc networks using calculation of New Retransmission Time out (RTO), to improve performance in term of congestion control. To calculate New RTO, adaptive backoff response approach (ABRA) in TCP New Reno was applied which suggest ABRA New Reno. It utilizes an ABRA by which congestion window and slow start threshold values were decreased whenever an acknowledgement is received and new backoff value calculate from smoothed round trip time. Evaluation based on comparative study of ABRA New Reno with other TCP Variants like New Reno and Reno was done using realistic parameters like TCP Packet Received, Packet Drop, Packets Retransmitted, Throughput, and Packet Delivery Ratio calculated by varying attributes of Node Speed, Number of Nodes and Pause Time. Implementation and simulations were performed in QualNet 4.0 simulator.

Analytical Study the Performance Evaluation of Mobile Ad Hoc Networks using AODV Protocol

The mobile Ad hoc networks (MANETs) have been communicating over wireless mesh channels and support multihop networks. Every node in the MANET is free to move independently in any of direction. MANET is a kind of wireless mesh networks which operates in Link Layer (LL) of Ad hoc networks. The current network scenarios consist of combined wired and wireless links which demands for investigating the performance in such mesh scenarios. This Paper investigates the performance evaluation of MANET over 802.11 wireless networks. MANET also uses TCP/IP structure to provide communication between different moving nodes. The impact of mobility or ad hoc routing protocols and different versions of TCP behave differently in this mobile environment and result obtained thereof has been presented in the form of important TCP metric. This paper further investigates the packet sequence traces in the wireless networks. We also use AODV routing protocols, perform best and less overhead in large networks. The different tools have been used to show the important TCP metrics such as Gnuplot, Perl and Python scripts.

Using simulation to test formally verified protocols in complex environments

Petri net modeling enables us to verify the protocol of interest formally. However, aside from formal verification, a new protocol really needs to be tested in a relatively realistic environment in which it interworks (or at least co-exists) with earlier or different versions of the same or similar protocols. Simulation excels in meeting such challenges. But the correctness of a protocol can never be proved by simulation alone. In this paper, we present an innovative methodology that combines the use of colored Petri nets and simulation (in ns-2) to obtain the advantages of deep formal verification with the broad spectrum testing of simulation. A new version of SACK TCP, α -min Paced SACK TCP, is used as the example protocol under test in our research. Our experimental results show that α -min Paced SACK TCP could reduce the number of packets queued at the bottleneck router significantly, decreasing the possibility of packet discard due to limited buffer space. Of course, the proposed methodology is a generic approach that can be used to study the performance of any new network protocol that is intended to run in an existing, complex network environment.