Explicit Congestion Notification Research Articles

Congestion‐free routing strategy in software defined data center networks

SummaryLarge‐scale online services and distributed execution engines (i.e., MapReduce and Dryad) generate large volumes of traffic in data center networks. As a consequence, significant congestion can occur in the data center network. To the best of our knowledge, most existing approaches either focus on local congestion‐aware mechanisms, which have only a poor ability to handle asymmetry or use explicit congestion notification packets, which are difficult to implement directly in switch hardware. These methods are insufficient to solve the congestion problem. In this paper, we focus on a congestion‐free routing strategy, resorting to the global view of the data center network in a software‐defined networking controller. Specifically, a timeslot allocation was first conducted for the coming packets, and then the corresponding routing paths were computed for each packet. In view of the efficiency, the timeslot allocation algorithm follows a heuristic pattern, and the path selection is modeled as a bin‐packing problem. Simulation results showed that the congestion‐free routing strategy proposed here performs well in throughput, queuing, and end‐to‐end round‐trip time. Copyright © 2015 John Wiley & Sons, Ltd.

Fast fairness convergence through fair rate estimation in Variable-structure congestion Control Protocol

Traditional Transmission Control Protocol (TCP) faces significant limitations such as unclear congestion implication, low utilization in high-speed networks, unstable throughput and limited fairness. In order to overcome such limitations, numerous research works have been done in effective congestion control algorithms. Among these, Variable-structure congestion Control Protocol (VCP) leverages only the existing two explicit congestion notification bits for network congestion feedback, yet achieves high utilization, low persistent queue length, negligible packet loss rate and reasonable fairness. However, VCP converges to fairness relatively slowly due to its large multiplicative decrease parameter. To address this problem, we propose an end-host based method for fast fairness convergence, namely VCP-FFC (VCP with fast fairness convergence). In each Additive Increase and Multiplicative Decrease (AIMD) epoch, VCP-FFC estimates a fair rate in end-hosts, and adjusts the congestion window according to the fair rate to achieve fast fairness convergence. We evaluate the performance of VCP-FFC over a wide range of network scenarios using NS2. Simulation Results show that VCP-FFC not only maintains the good properties of VCP, but also significantly accelerates fairness convergence. Dynamic analysis shows that VCP-FFC is asymptotically stable and converges to a unique stationary point, i.e. the fair rate share.

A Survey on Congestion Notification Algorithm in Data Centers

days, due to fast exploitation of modern data center networks (DCNs), Ethernet is most prominently used than the conventional storage networking technologies like Fiber Channel and Infiniband. Low-loss, low-delay operations are key aspect of these conventional technologies that make them suitable for datacenter. Accordingly IEEE 802.1 standards committee is developing new specification for congestion management for Ethernet in datacenter networks. Backward Congestion Notification (BCN), Forward Explicit Congestion Notification (FECN), Enhanced Forward Explicit Congestion Notification (E-FECN), Quantizes Congestion Notification (QCN), Approximate Fair Quantizes Congestion Notification (AF-QCN) and Fair Quantizes Congestion Notification (F- QCN) is six proposals. Each of the proposals has its own advantages and disadvantages. In this paper, we provide a comprehensive survey of the most recent research activities in DCNs, with emphasis on the network congestion notification algorithms. We present a brief overview of the congestion problem along with previously proposed solutions.

Primary Voltage Control in Active Distribution Networks via Broadcast Signals: The Case of Distributed Storage

In this paper, we consider an active distribution network (ADN) that performs primary voltage control using real-time demand response via a broadcast low-rate communication signal. The ADN also owns distributed electrical energy storage. We show that it is possible to use the same broadcast signal deployed for controlling loads to manage the distributed storage. To this end, we propose an appropriate control law to be embedded into the distributed electrical storage controllers that reacts to the defined broadcast signal in order to control both active and reactive power injections. We analyze, in particular, the case where electrical storage systems consist of supercapacitor arrays and where the ADN uses the grid explicit congestion notification (GECN) for real-time demand response that the authors have developed in a previous contribution. We estimate the energy reserve required for successfully performing voltage control depending on the characteristics of the network. The performance of the scheme is numerically evaluated on the IEEE 34-node test feeder. We further evaluate the effect, depending on the line characteristics, of reactive versus active power controlled injections. We find that without altering the demand-response signal, a suitably designed controller implemented in the storage devices enables them to successfully contribute to primary voltage control.

FaST: Fine-grained and Scalable TCP for Cloud Data Center Networks

With the increasing usage of cloud applications such as MapReduce and social networking, the amount of data traffic in data center networks continues to grow. Moreover, these appli-cations follow the incast traffic pattern, where a large burst of traffic sent by a number of senders, accumulates simultaneously at the shallow-buffered data center switches. This causes severe packet losses. The currently deployed TCP is custom-tailored for the wide-area Internet. This causes cloud applications to suffer long completion times towing to the packet losses, and hence, results in a poor quality of service. An Explicit Congestion Notification (ECN)-based approach is an attractive solution that conservatively adjusts to the network congestion in advance. This legacy approach, however, lacks scalability in terms of the number of flows. In this paper, we reveal the primary cause of the scalability issue through analysis, and propose a new congestion-control algorithm called FaST. FaST employs a novel, virtual congestion window to conduct fine-grained congestion control that results in improved scalability. Fur-thermore, FaST is easy to deploy since it requires only a few software modifications at the server-side. Through ns-3 simulations, we show that FaST improves the scalability of data center networks compared with the existing approaches.

GECN: Primary Voltage Control for Active Distribution Networks via Real-Time Demand-Response

Demand response (DR) has traditionally targeted peak shaving for the optimal allocation of electricity consumption on a time scale that ranges from minutes to hours. However, with the availability of advanced monitoring and communication infrastructure, the potential of real-time DR for providing ancillary services to the grid has not yet been adequately explored. In this work, we propose a low-overhead decentralized DR control mechanism, henceforth called Grid Explicit Congestion Notification (GECN), intended for deployment by distribution network operators (DNOs) to provide ancillary services to the grid by a seamless control of a large population of elastic appliances. Contrary to classic DR approaches, the proposed scheme aims to continuously support the grid needs in terms of voltage control by broadcasting low-bit rate control signals on a fast time scale (i.e., every few seconds). Overall, the proposed DR mechanism is designed to i) indirectly reveal storage capabilities of end-customers and ii) have a negligible impact on the end-customer. In order to estimate the benefits of the proposed mechanism, the evaluation of the algorithm is carried out by using the IEEE 13 nodes test feeder in combination with realistic load profiles mixed with non-controllable demand and non-dispatchable generation from photovoltaic distributed generation.

An Ecn Approach to Congestion Control Mechanisms in Mobile Adhoc Networks

Node(s)/link(s) of a network are subjected to overloading; network performance deteriorates substantially due to network congestion. Network congestion can be mitigated with the help of Explicit Congestion notification (ECN) technique. ECN notification is carried out by setting ECN bit in the TCP header. This allows for end-to-end notification of network congestion without dropping packets. ECN bit notifies TCP sources of incipient congestion before losses occur. ECN is a binary indicator (1 bit) which does not reflect the congestion level completely and so its convergence speed is relatively low. In our work, we have used an extra ECN bit (2 bit ECN). The extra bit allows for passing of additional congestion feedback to the source node. This enables the source node to determine the level of congestion based on which steps can be taken to ensure faster convergence. In comparison to single bit ECN, the additional information afforded by double bit ECN allows for more flexibility to adjust window size, to handle congestion. Simulation results have shown that the proposed method improves overall performance of the network by over 12%.

Kalman Filter Applied to a Active Queue Management Problem

The Kalman Filter (KF) is used to optimally estimate system states from the sequential noisy measurements of the outputs. On the other hand, real-time systems are often modeled with uncertainties and time-delays. Developing KF algorithms for such systems is an important problem to obtain optimal state estimates by utilizing the information on uncertainties and time-delays. First, designing of KF for nominal discrete-time systems is studied. Considering the covariance of the error in the estimation the KF algorithm is derived which is further tested on a numerical example. A designed KF is tested on a real-time problem i.e. Active Queue Management (AQM) problem.In Internet routers, Active Queue Management (AQM) is a technique that consists in dropping or Explicit Congestion Notification (ECN) marking packets before a router's queue is full. Keywords: Active Queue Management,Explicit Congestion Notification, discrete time system, time-delay.

Max-min fair rate allocation for VCP in multi-bottleneck networks

Variable-structure congestion Control Protocol VCP leverages the existing two Explicit Congestion Notification ECN bits for network congestion feedback and achieves comparable performance to eXplicit congestion Control Protocol XCP. VCP utilises the encoded load factor-based feedback scheme to deliver congestion signals. In multi-bottleneck networks, all bottleneck links on the transmission path would contribute to the aggregate congestion signal, and VCP fails to achieve max-min fairness. To address this problem, a new metric called overload period is introduced to measure the congestion level of links, and the overload period-based feedback scheme is proposed to make sure that each VCP flow's rate is controlled only by the link with minimum overload period encountered on the transmission path. It is proved that the link with minimum overload period is the flow's bottleneck, and the max-min fair rate allocation is achieved in multi-bottleneck networks. The overload period-based feedback scheme is validated using NS2 simulations.

Real-time detection of intrusive traffic in QoS network domains

A capable, scalable, and reliable model detects intrusive traffic by investigating the impact of user behavior on quality-of-service regulations in real time. The model also proposes reliable coordination for investigating user traffic, including traffic injected through several gateways. Traffic investigation is triggered only when the network is congested; at that moment, burst gateways generate an echo of explicit congestion notification to misbehaving users. The model investigates these users by measuring their bandwidth consumption ratios. User traffic that exceeds the service-level agreement bandwidth ratio is filtered as intrusive. Simulation results demonstrate that the proposed model efficiently monitors user behavior and detects intrusive traffic.

An Effective and Feasible Congestion Management Technique for High-Performance MINs with Tag-Based Distributed Routing

As parallel computing systems increase in size, the interconnection network is becoming a critical subsystem. The current trend in network design is to use as few components as possible to interconnect the end nodes, thereby reducing cost and power consumption. However, this increases the probability of congestion appearing in the network. As congestion may severely degrade network performance, the use of a congestion management mechanism is becoming mandatory in modern interconnects. One of the most cost-effective proposals to deal with the problems derived from congestion situations is the Regional Explicit Congestion Notification (RECN) strategy, based on using special queues to totally isolate the packet flows which contribute to congestion, thereby preventing the Head-of-Line (HoL) blocking effect that these flows may cause to others. Unfortunately, RECN requires the use of source-based routing, thus not being suitable for interconnects with distributed routing, like InfiniBand. Although some RECN-like mechanisms have been proposed for distributed-routing networks, they are not scalable due to the huge amount of control memory that they require in medium-size or large networks. In this paper, we propose Distributed-Routing-Based Congestion Management (DRBCM), a new scalable technique which, following the RECN principles, totally prevents congestion from producing HoL-blocking in multistage interconnection networks (MINs) using tag-based distributed routing. Simulation results indicate that, regardless of network size, DRBCM presents small resource requirements to keep network performance at maximum level even in scenarios of heavy congestion, where it utterly outperforms (with a gain up to 70 percent) current solutions for distributed-routing networks, like the InfiniBand congestion-control mechanism based on injection throttling. Thus, DRBCM is an efficient, cost-effective, and scalable solution for congestion management.

Performance Enhancement of Network Transmission by Change of Delay Parameter to Increase Throughput

Congestion is a major problem which really degrades network performance. Its effect on the delay is quite high which decrease the throughput in considerable amount. In this paper we have proposed a new congestion scheme which will enhance performance of network transmission by considering delay and throughput as main components. For this we are using Adaptive Explicit Congestion Notification method which allows us to control the network transmission from source to a receiver in a hierarchical manner by alterations in the receiving sending capacity. To achieve predictable average delays with adaptive explicit congestion notification would require constant tuning of the parameters to adjust to current traffic conditions. The sending and receiving ends are tuned according to its capacity and current network traffic condition once ECN (Explicit congestion notification) notification packet comes into action. The aim of this paper is to solve the parameter tuning problem of the AECN by dynamically setting up the network parameters to overcome delay in network transmission and hence to increase throughput .We will be comparing the performance of the ECN enabled system with the AECN enabled system. For this we are going to use JAVA , JNETPCAP and WINPCAP API’s by which we can create TCP packets, alter the fields of headers, send and receive packets etc.

Easing Congestion in Computer Networks using the Receiver-Window Modification (RWM) Scheme

Random Early Detection (RED), Adaptive RED (ARED), BLUE and other Active Queue Management (AQM) queues have been proposed to replace drop-tail queues for Transmission Control Protocol/Internet Protocol (TCP/IP) networks to ease network congestion. Together with the Exp licit Congestion Notification (ECN) scheme, they had shown some pro mise over the drop tail queues. However, the timeout mechanism or the duration of the reception of three duplicate acknowledgements (ACKs), due to early-dropped packets, delays the response time of TCP in reducing the network congestion. Moreover, using ECN has its downsides: i) its messages may get lost; and ii) TCP implementations at the source, router and the destination have to be ECN-co mpliant (which presents a significant problem in today's implementations). To minimize these problems, this paper presents a novel AQM modification scheme called Receiver-W indow Modification (RWM), wh ich can be used together with any AQM queue for congestion avoidance in packet switched networks, especially at ingress and gateway routers. RWM improves on the average queue sizes, one-way end-to-end packet delays, delay jitter, throughput and number of dropped packets. It also overcomes the dependency of these AQM queues on the queues of downstream routers since ECN messages may get dropped or delayed due to congestion at these routers. We carry out extensive NS2 simulat ions to show our results and to support our claims.

Congestion Control With Multipacket Feedback

Many congestion control protocols use explicit feedback from the network to achieve high performance. Most of these either require more bits for feedback than are available in the IP header or incur performance limitations due to inaccurate congestion feedback. There has been recent interest in protocols that obtain high-resolution estimates of congestion by combining the explicit congestion notification (ECN) marks of multiple packets, and using this to guide multiplicative increase, additive increase, multiplicative decrease (MI-AI-MD) window adaptation. This paper studies the potential of such approaches, both analytically and by simulation. The evaluation focuses on a new protocol called Binary Marking Congestion Control (BMCC). It is shown that these schemes can quickly acquire unused capacity, quickly approach a fair rate distribution, and have relatively smooth sending rates, even on high bandwidth-delay product networks. This is achieved while maintaining low average queue length and negligible packet loss. Using extensive simulations, we show that BMCC outperforms XCP, VCP, MLCP, CUBIC, CTCP, SACK, and in some cases RCP, in terms of average flow completion times. Suggestions are also given for the incremental deployment of BMCC.

A Novel Cross-layer Congestion Control Strategy for Wireless Sensor Networks

Network congestion in wireless sensor networks (WSNs) leads to packet drops, throughput reduction and degradation of channel quality which in turn decrease network performance. A novel cross-layer congestion control strategy for WSNs is presented. The strategy performs the bandwidth and delay estimation in MAC layer for the current link, and feedbacks them to the transport layer through the cross-layer interaction mechanism. The transport layer of source nodes adjusts the congestion window and slow start threshold based on the received information of path bandwidth and delay and Explicit Congestion Notification (ECN), for mitigating network congestion. The simulation results show that the strategy can improve the utilization of network resources, reduce unnecessary packets retransmission, and significantly improve the performance of WSNs.

Extended Queue Management Backward Congestion Control Algorithms for Web Transfers in Wireless Environment

Wireless links are characterized by high error rates and intermittent connectivity. TCP congestion control has been developed on the assumption that network congestion is the only cause for packet loss. Upon detecting a packet loss, TCP drops its transmit window resulting in an unnecessary reduction of end-to-end throughput which results in suboptimal performance.The sender has to be made aware by some feedback mechanism that some of the losses reported are not due to congestion. The Active Queue Management algorithms (AQM) are used to reduce congestion, and in this paper, we have analysed four AQM algorithms, Random Early Deduction (RED), Wireless Explicit Congestion Notification (WECN), Queue Management Backward Congestion Control Algorithm (QMBCCA) and its enhanced version Extended Queue Management Backward Congestion Control Algorithm (EQMBCCA). WECN, QMBCCA & EQMBCCA algorithms make use of feedback mechanisms. WECN gives feedback using the CE bit. QMBCCA and EQMBCCA make use of ISQ notifications and also the CE bit whenever the average queue size crosses minimum threshold value. EQMBCCA reduces the reverse ISQ traffic by introducing a configurable intermediate threshold value IntThres. The comparison is made in terms of Delay, HTTP packet loss percentage and fairness for FTP flows in a wireless environment. It is found that the performance of EQMBCCA is almost equal to that of QMBCCA and better than RED and WECN.

Progressive Congestion Management Based on Packet Marking and Validation Techniques

Congestion management in multistage interconnection networks is a serious problem, which is not solved completely. In order to avoid the degradation of network performance when congestion appears, several congestion management mechanisms have been proposed. Most of these mechanisms are based on explicit congestion notification. For this purpose, switches detect congestion and depending on the applied strategy, packets are marked to warn the source hosts. In response, source hosts apply some corrective actions to adjust their packet injection rate. Although these proposals seem quite effective, they either exhibit some drawbacks or are partial solutions. Some of them introduce some penalties over the flows not responsible for congestion, whereas others can cope only with congestion situations that last for a short time. In this paper, we present an overview of the different strategies to detect and correct congestion in multistage interconnection networks, and propose a new mechanism referred to as Marking and Validation Congestion Management (MVCM), targeted to this kind of lossless networks, and based on a more refined packet marking strategy combined with a fair set of corrective actions, that makes the mechanism able to effectively manage congestion regardless of the congestion degree. Evaluation results show the effectiveness and robustness of the proposed mechanism.

A Transport Protocol to Exploit Multipath Diversity in Wireless Networks

Wireless networks (including wireless mesh networks) provide opportunities for using multiple paths. Multihoming of hosts, possibly using different technologies and providers, also makes it attractive for end-to-end transport connections to exploit multiple paths. In this paper, we propose a multipath transport protocol, based on a carefully crafted set of enhancements to TCP, that effectively utilizes the available bandwidth and diversity provided by heterogeneous, lossy wireless paths. Our Multi-Path LOss-Tolerant (MPLOT) transport protocol can be used to obtain significant goodput gains in wireless networks, subject to bursty, correlated losses with average loss rates as high as 50%. MPLOT is built around the principle of separability of reliability and congestion control functions in an end-to-end transport protocol. Congestion control is performed separately on individual paths, and the reliability mechanism works over the aggregate set of paths available for an end-to-end session. MPLOT distinguishes between congestion and link losses through Explicit Congestion Notification (ECN), and uses Forward Error Correction (FEC) coding to recover from data losses. MPLOT uses a dynamic packet mapping based on the current path characteristics to choose a path for a packet. Use of erasure codes and block-level recovery ensures that in MPLOT the receiving transport entity can recover all data as long as a necessary number of packets in the block are received, irrespective of which packets are lost. We present a theoretical analysis of the different design choices of MPLOT and show that MPLOT chooses its policies and parameters such that a desirable tradeoff between goodput with data recovery delay is attained. We evaluate MPLOT, through simulations, under a variety of test scenarios and demonstrate that it effectively exploits path diversity in addition to efficiently aggregating path bandwidths while remaining fair to a conventional TCP flow on each path.

A cognitive accountability mechanism for penalizing misbehaving ECN‐based TCP stacks

SUMMARYThe introduction of high‐bandwidth demanding services such as multimedia services has resulted in important changes on how services in the Internet are accessed and what quality‐of‐experience requirements (i.e. limited amount of packet loss, fairness between connections) are expected to ensure a smooth service delivery. In the current congestion control mechanisms, misbehaving Transmission Control Protocol (TCP) stacks can easily achieve an unfair advantage over the other connections by not responding to Explicit Congestion Notification (ECN) warnings, sent by the active queue management (AQM) system when congestion in the network is imminent. In this article, we present an accountability mechanism that holds connections accountable for their actions through the detection and penalization of misbehaving TCP stacks with the goal of restoring the fairness in the network. The mechanism is specifically targeted at deployment in multimedia access networks as these environments are most prone to fairness issues due to misbehaving TCP stacks (i.e. long‐lived connections and a moderate connection pool size). We argue that a cognitive approach is best suited to cope with the dynamicity of the environment and therefore present a cognitive detection algorithm that combines machine learning algorithms to classify connections into well‐behaving and misbehaving profiles. This is in turn used by a differentiated AQM mechanism that provides a different treatment for the well‐behaving and misbehaving profiles. The performance of the cognitive accountability mechanism has been characterized both in terms of the accuracy of the cognitive detection algorithm and the overall impact of the mechanism on network fairness. Copyright © 2012 John Wiley & Sons, Ltd.

Adaptive active queue management controller for TCP communication networks using PSO-RBF models

Addressing performance degradations in end-to-end congestion control has been one of the most active research areas in the last decade. Active queue management (AQM) aims to improve the overall network throughput, while providing lower delay and reduce packet loss and improving network. The basic idea is to actively trigger packet dropping (or marking provided by explicit congestion notification (ECN)) before buffer overflow. Radial bias function (RBF)-based AQM controller is proposed in this paper. RBF controller is suitable as an AQM scheme to control congestion in TCP communication networks since it is nonlinear. Particle swarm optimization (PSO) algorithm is also employed to derive RBF parameters such that the integrated-absolute error (IAE) is minimized. Furthermore, in order to improve the robustness of RBF controller, an error-integral term is added to RBF equation. The results of the comparison with Drop Tail, adaptive random early detection (ARED), random exponential marking (REM), and proportional-integral (PI) controllers are presented. Integral-RBF has better performance not only in comparison with RBF but also with ARED, REM and PI controllers in the case of link utilization while packet loss rate is small.