Source-ordering for improved TCP performance over load-balanced optical burst-switched (OBS) networks

Recent advances in optical switching technology allow for the creation of networks in which data bursts are switched optically at each node, offering a greater degree of flexibility suitable for handling bursty Internet traffic. TCP-based applications account for a majority of data traffic in the Internet; thus understanding and improving the performance of TCP implementations over OBS networks are critical. Previously, several articles show that load-balanced routing improves loss-performance in OBS. In this paper, we identify the ill-effects of load-balanced OBS on TCP performance caused by false time-outs and false fast-retransmits. We propose a source-ordering mechanism that significantly improves TCP throughput over a load-balanced OBS network.     

光突发交换网络的动态TCP特性研究

通过仿真研究了动态业务竞争环境下,不同TCP版本在OBS网络中的传输性能。研究中,以TCP在OBS网络中传送一个固定大小的文件所需的时间作为性能指标,此外,详细分析了突发丢失的内部机制和造成不同TCP版本性能差异的原因。结果表明,TCP SACK性能最好。     

TCP over optical burst-switched networks with controlled burst retransmission

For optical burst-switched (OBS) networks in which TCP is implemented at a higher layer, the loss of bursts can lead to serious degradation of TCP performance. Due to the bufferless nature of OBS, random burst losses may occur, even at low traffic loads. Consequently, these random burst losses may be mistakenly interpreted by the TCP layer as congestion in the network. The TCP sender will then trigger congestion control mechanisms, thereby reducing TCP throughput unnecessarily. In this paper, we introduce a controlled retransmission scheme in which the bursts lost due to contention in the OBS network are retransmitted at the OBS layer. The OBS retransmission scheme can reduce the burst loss probability in the OBS core network. Also, the OBS retransmission scheme can reduce the probability that the TCP layer falsely detects congestion, thereby improving the TCP throughput. We develop an analytical model for evaluating the burst loss probability in an OBS network that uses a retransmission scheme, and we also analyze TCP throughput when the OBS layer implements burst retransmission. We develop a simulation model to validate the analytical results. Simulation and analytical results show that an OBS layer with controlled burst retransmission provides up to two to three orders of magnitude improvement in TCP throughput over an OBS layer without burst retransmission. This significant improvement is primarily because the TCP layer triggers fewer time-outs when the OBS retransmission scheme is used.     

Multi-layer loss recovery in TCP over optical burst-switched networks

It is well-known that the bufferless nature of optical burst-switching (OBS) networks cause random burst loss even at low traffic loads. When TCP is used over OBS, these random losses make the TCP sender decrease its congestion window even though the network may not be congested. This results in significant TCP throughput degradation. In this paper, we propose a multi-layer loss-recovery approach with automatic retransmission request (ARQ) and Snoop for OBS networks given that TCP is used at the transport layer. We evaluate the performance of Snoop and ARQ at the lower layer over a hybrid IP-OBS network. Based on the simulation results, the proposed multi-layer hybrid ARQ + Snoop approach outperforms all other approaches even at high loss probability. We developed an analytical model for end-to-end TCP throughput and verified the model with simulation results.     

Improving performance of TCP over mobile wireless networks

Mobile IP is a network layer protocol for handling mobility of hosts in the Internet. However, mobile IP handoff causes degradation of TCP performance. Hence, there is a need for improving performance of TCP over mobile IP in wireless mobile networks. We propose an approach which handles losses due to both wireless link errors and host mobility. To handle losses due to host mobility, a method for seamless handoff is proposed. Empirical results show that the scheme provides substantial improvement of performance.

Reducing the impact of false time out on TCP performance in TCP over OBS networks

Random burst contention losses plague the performance of Optical Burst Switched networks. Such random losses occur even in low load network condition due to the analogous behavior of wavelength and routing algorithms. Since a burst may carry many packets from many TCP sources, its loss can trick the TCP sources to conclude/infer that the underlying (optical) network is congested. Accordingly, TCP reduces sending rate and switches over to either fast retransmission or slow start state. This reaction by TCP is uncalled-for in TCP over OBS networks as the optical network may not be congested during such random burst contention losses. Hence, these losses are to be addressed in order to improve the performance of TCP over OBS networks. Existing work in the literature achieves the above laid objective at the cost of violating the semantics of OBS and/or TCP. Several other works make delay inducing assumptions. In our work, we introduce a new layer, called Adaptation Layer, in between TCP and OBS layers. This layer uses burst retransmission to mitigate the effect of burst loss due to contention on TCP by leveraging the difference between round trip times of TCP and OBS. We achieve our objective with the added advantage of maintaining the semantics of the layers intact.     

TCP over OBS边缘节点重传机制的研究

使用传输控制协议(TCP)的拥塞控制机制在传统互联网技术中是解决问题的很好方法.然而在面对新型的光网络技术OBS网络,简单地采用TCP技术因受到OBS自身特性的影响,性能表现不佳,吞吐量低下.为此,通过对OBS技术和TCP拥塞控制机制深入的研究,提出一种基于OBS边缘节点的ACK重传机制来提升TCP over OBS性能的方法,提高OBS网络吞吐量.     

Improving TCP performance over wireless networks at the link layer

We present the transport unaware link improvement protocol (TULIP), which dramatically improves the performance of TCP over lossy wireless links, without competing with or modifying the transport- or network-layer protocols. TULIP is tailored for the half-duplex radio links available with today's commercial radios and provides a MAC acceleration feature applicable to collision-avoidance MAC protocols (e.g., IEEE 802.11) to improve throughput. TULIP's timers rely on a maximum propagation delay over the link, rather than performing a round-trip time estimate of the channel delay. The protocol does not require a base station and keeps no TCP state. TULIP is exceptionally robust when bit error rates are high; it maintains high goodput, i.e., only those packets which are in fact dropped on the wireless link are retransmitted and then only when necessary. The performance of TULIP is compared against the performance of the Snoop protocol (a TCP-aware approach) and TCP without link-level retransmission support. The results of simulation experiments using the actual code of the Snoop protocol show that TULIP achieves higher throughput, lower packet delay, and smaller delay variance. This revised version was published online in June 2006 with corrections to the Cover Date.     

Improving TCP performance over wireless networks at the link layer

We present the transport unaware link improvement protocol (TULIP), which dramatically improves the performance of TCP over lossy wireless links, without competing with or modifying the transport- or network-layer protocols. TULIP is tailored for the half-duplex radio links available with today's commercial radios and provides a MAC acceleration feature applicable to collision-avoidance MAC protocols (e.g., IEEE 802.11) to improve throughput. TULIP's timers rely on a maximum propagation delay over the link, rather than performing a round-trip time estimate of the channel delay. The protocol does not require a base station and keeps no TCP state. TULIP is exceptionally robust when bit error rates are high; it maintains high goodput, i.e., only those packets which are in fact dropped on the wireless link are retransmitted and then only when necessary. The performance of TULIP is compared against the performance of the Snoop protocol (a TCP-aware approach) and TCP without link-level retransmission support. The results of simulation experiments using the actual code of the Snoop protocol show that TULIP achieves higher throughput, lower packet delay, and smaller delay variance. This revised version was published online in June 2006 with corrections to the Cover Date.     

Improving TCP performance in mobile networks

In this letter, a new transport layer mechanism is proposed to improve the performance of transport control protocol (TCP) in mobile networks. The proposed mechanism is comprised of two parts: a loss classifier (LC) and a congestion window extrapolator (CWE). Based on LC, the cause of packet loss during roaming is determined. If the loss is considered to be caused by congestion in the wireline, the congestion window is halved; otherwise, the packet is considered to be lost in the last hop, the wireless portion, and the sender adjusts the size of the congestion window based on CWE. We conduct simulations to evaluate the performance of the proposed mechanism. The results show that our mechanism significantly improves TCP performance as compared with existing solutions for mobile networks.     

Improving TCP performance over networks with wireless components using ‘probing devices’

TCP error control mechanism lacks the ability to detect with precision the nature of potential errors during communication. It is only capable of detecting the results of the errors, namely that segments are dropped. As a result, the protocol lacks the ability to implement an appropriate error recovery strategy cognizant of current network conditions and responsive to the distinctive error characteristics of the communication channel. TCP sender always calls for the sending window to shrink. We show that probing mechanisms could enhance the error detection capabilities of the protocol. TCP could then flexibly adjust its window in a manner that permits the available bandwidth to be exploited without violating the requirements of stability, efficiency and fairness that need to be guaranteed during congestion. Our experiments have three distinct goals: First, to demonstrate the potential contribution of probing mechanisms. A simple probing mechanism and an immediate recovery strategy are grafted into TCP‐Tahoe and TCP‐Reno. We show that, this way, standard TCP can improve its performance without requiring any further change. Second, to study the performance of adaptive strategies. An adaptive TCP with probing is used, that is responsive to the detected error conditions by alternating slow start, fast recovery and immediate recovery. An adaptive error recovery strategy can yield better performance. Third, to study the design limitations of the probing device itself. The aggressive or conservative nature of the probing mechanisms themselves can determine the aggressive or conservative behaviour of the protocol and exploit accordingly the energy/throughput trade‐off. Copyright © 2002 John Wiley & Sons, Ltd.     

Using multiple per egress burstifiers for enhanced TCP performance in OBS networks

Burst assembly mechanism is one of the fundamental factors that determine the performance of an optical burst switching (OBS) network. In this paper, we investigate the influence of the number of burstifiers on TCP performance for an OBS network. The goodput of TCP flows between an ingress node and an egress node traveling through an optical network is studied as the number of assembly buffers per destination varies. First, the burst-length independent losses resulting from the contention in the core OBS network using a non-void-filling burst scheduling algorithm, e.g., Horizon, are studied. Then, burst-length dependent losses arising as a result of void-filling scheduling algorithms, e.g., LAUC-VF, are studied for two different TCP flow models: FTP-type long-lived flows and variable size short-lived flows. Simulation results show that for both types of scheduling algorithms, both types of TCP flow models, and different TCP versions (Reno, Newreno and Sack), TCP goodput increases as the number of burst assemblers per egress node is increased for an OBS network employing timer-based assembly algorithm. The improvement from one burstifier to moderate number of burst assemblers is significant (15–50% depending on the burst loss probability, per-hop processing delay, and the TCP version), but the goodput difference between moderate number of buffers and per-flow aggregation is relatively small, implying that an OBS edge switch should use moderate number of assembly buffers per destination for enhanced TCP performance without substantially increasing the hardware complexity.

Efficient burst scheduling algorithms in optical burst-switched networks using geometric techniques

Optical burst switching (OBS) is a promising paradigm for the next-generation Internet. In OBS, a key problem is to schedule bursts on wavelength channels, whose bandwidth may become fragmented with the so-called void (or idle) intervals, using both fast and bandwidth efficient algorithms so as to reduce burst loss. To date, two well-known scheduling algorithms, called Horizon and LAUC-VF, have been proposed in the literature, which trade off bandwidth efficiency for fast running time and vice versa, respectively. In this paper, we propose a set of novel burst scheduling algorithms for OBS networks with and without fiber delay lines (FDLs) utilizing the techniques from computational geometry. In networks without FDLs, our proposed minimum-starting-void (Min-SV) algorithm can schedule a burst in O(logm) time, where m is the total number of void intervals, as long as there is a suitable void interval. Simulation results suggest that our algorithm achieves a loss rate which is at least as low as LAUC-VF, but can run much faster. In fact, its speed can be almost the same as Horizon (which has a much higher loss rate). In networks with FDLs, our proposed batching FDL algorithm considers a batch of FDLs to find a suitable FDL to delay a burst which would otherwise be discarded due to contention, instead of considering the FDLs one by one. The average running time of this algorithm is therefore significantly reduced from that of the existing burst scheduling algorithms. Our algorithms can also be used as algorithmic tools to speed up the scheduling time of many other void-filling scheduling algorithms.     

Improving Reno and New-Reno's performances over OBS networks without SACK

In TCP over optical burst switching (OBS) networks, consecutive multiple packet losses are common since an optical burst usually contains a number of consecutive packets from the same TCP sender. It has been proved that over OBS networks Reno and New-Reno achieve lower throughput performances than that of SACK, which can address the inefficiency of Reno and New-Reno in dealing with consecutive multiple packet losses. However, SACK adopts complex mechanisms not only at the sender's but also at the receiver's protocol stack, and thus has a higher difficulty in deployment.In this paper we propose B-Reno, a new TCP implementation designed for TCP over OBS networks. Using some simple modifications to New-Reno only at the sender's protocol stack, B-Reno can overcome the inefficiencies of Reno and New-Reno in dealing with consecutive multiple packet losses and thus improve their throughputs over OBS networks. Moreover, B-Reno can also achieve performance similar with that of SACK over OBS networks while avoiding SACK's difficulty in deployment due to complex mechanisms at both the sender's and the receiver's protocol stack.     

A new fault-management method using congestion-avoidance routing for optical burst-switched networks

Optical burst switching (OBS) has been proposed as a promising technique to support high-bandwidth, bursty data traffic in the next-generation optical Internet. This study investigates a new fault-management framework for an OBS network. In an OBS network, burst-loss performance is a critical concern. In OBS, the data-burst transmission is delayed by an offset time (relative to its burst control packet (BCP), or header), and the burst follows its header without waiting for an acknowledgment for resource reservation. Thus, a burst may be lost at an intermediate node due to contention, which is generally resolved according to the local routing and bandwidth information. The routing table maintained in each OBS node is generally pre-computed and fixed to forward the data bursts. Such a static forwarding feature might have limited efficiency to resolve contentions. Moreover, a burst may be lost and the network may be congested when a network element (e.g., fiber link) fails. Therefore, for reliable burst transport, we develop dynamic routing approaches for preplanned congestion avoidance. Our goal is to proactively avoid congestion during the route-computation process, and our approach employs wavelength-channel utilization, traffic distribution, and link-distance information in the proposed objective functions for routing. Two update mechanisms for maintaining a dynamic routing table are presented to accommodate bursty data traffic. Based on our routing mechanisms, we also propose a new congestion-avoidance-and-protection (CAP) approach, which employs a primary route and a group of backup routes for each node pair against failures and congestion. The performance of the proposed protection strategy using congestion-avoidance routing is demonstrated to be promising through illustrative numerical examples.

Intermediate-node-initiation (INI): A generalized signaling framework for optical burst-switched networks

In this paper, we discuss different signaling techniques for optical burst-switched networks. We develop a generalized signaling framework for optical burst-switched networks, which provides guidelines about the performance of each signaling technique based on the different parameters in the framework. The two commonly used signaling techniques in optical burst switching are two-way based tell-and-wait (TAW) and one-way based just-enough-time (JET). TAW suffers from high end-to-end packet delay, while JET suffers from high packet loss. There is no signaling technique that offers flexibility in terms of both loss and delay. We propose a hybrid signaling technique called intermediate-node-initiated (INI) signaling for optical burst-switched networks. INI can provide different levels of loss and delay characteristics based on end-user application requirements. The granularity of INI ranges between the one-way based and the two-way based signaling techniques. In INI reservation of channels is initiated at an intermediate node, known as the initiating node, in both forward and backward directions at the same time. We show that by appropriately selecting the initiating node, we can simulate both TAW and JET using the INI signaling technique. Through simulations, we shown that INI performs better than TAW in terms of average end-to-end packet delay and better than JET in terms of burst loss probability. We extend the INI signaling technique to provide QoS differentiation in the OBS core, differentiated INI (DINI), by carefully choosing different initiation nodes depending on delay and loss requirements of end-user applications. Through extensive simulations, we show that the DINI technique outperforms the existing offset-based QoS technique.     

Improving TCP performance in ad hoc networks using signal strength based link management

Mobility in ad hoc networks causes frequent link failures, which in turn causes packet losses. TCP attributes these packet losses to congestion. This incorrect inference results in frequent TCP re-transmission time-outs and therefore a degradation in TCP performance even at light loads. We propose mechanisms that are based on signal strength measurements to alleviate such packet losses due to mobility. Our key ideas are (a) if the signal strength measurements indicate that a link failure is most likely due to a neighbor moving out of range, in reaction, facilitate the use of temporary higher transmission power to keep the link alive and, (b) if the signal strength measurements indicate that a link is likely to fail, initiate a route re-discovery proactively before the link actually fails. We make changes at the MAC and the routing layers to predict link failures and estimate if a link failure is due to mobility. We also propose a simple mechanism at the MAC layer that can help alleviate false link failures, which occur due to congestion when the IEEE 802.11 MAC protocol is used. We compare the above proactive and reactive schemes and also demonstrate the benefits of using them together and along with our MAC layer extension. We show that, in high mobility, the goodput of a TCP session can be improved by as much as 75% at light loads (when there is only one TCP session in the network) when our methods are incorporated. When the network is heavily loaded (i.e., there are multiple TCP sessions in the network), the proposed schemes can improve the aggregate goodput of the TCP sessions by about 14–30%, on average.     

Improving TCP performance for vertical handover in heterogeneous wireless networks

The incorporation of wireless local area networks (WLANs) into existing cellular networks as supplementary access technologies has become an issue of great interest. However, vertical handover (VHO), which allows users to roam between a WLAN and a cellular network, causes an abrupt change in certain link characteristics such as the round trip time and data rate. Owing to such changes, reordering problem and premature timeout occur and trigger unnecessarily fast retransmission during VHO, causing throughput degradation. Thus, we propose a new transmission control protocol (TCP) mechanism, which resolves the reordering problem by suppressing unnecessary retransmission caused by spurious duplicate acknowledgments (dupacks) incurred because of the reordering problem, and prevents premature timeout by employing an adaptive retransmission timer. We analytically investigate the throughput of our proposed TCP scheme. The numerical and simulation results show that our proposed TCP performs better in terms of throughput than other schemes appearing in the literature. Copyright © 2009 John Wiley & Sons, Ltd.     

Improving throughput of long-hop TCP connections in IP over OPS networks

Optical Packet Switching (OPS) can provide the ever-increasing bandwidth required for Internet traffic and new applications for future networks. However, optical packet loss is the major problem for an OPS network. Moreover, by increasing the number of hops between a pair of ingress–egress switches in an OPS network, optical Packet Loss Rate (PLR) between this pair is increased. Therefore, due to a higher PLR for long-hop TCP connections, the throughput of these connections may be much lower than the short-hop TCP connections. To overcome this problem, it is proposed in this paper to use the retransmission idea in the optical domain not only to increase TCP throughput but also to improve the throughput of multi-hop TCP connections, and also to have a loss-free OPS network. Under retransmission in the optical domain, a copy of the transmitted traffic is kept in the electronic buffers of ingress switches and retransmitted in the optical domain whenever required. Note that the TCP layer has its own retransmission at the client packet level as well. By retransmission of lost packets in the optical domain, TCP would be unaware of the lost client packets, and therefore, TCP would not reduce its sending rate. In this paper, TCP throughput is studied in a bufferless slotted OPS network and the effectiveness of the proposed mechanism is evaluated.

Understanding and improving TCP performance over networks withminimum rate guarantees

A large number of Internet applications are sensitive to overload conditions in the network. While these applications have been designed to adapt somewhat to the varying conditions in the Internet, they can benefit greatly from an increased level of predictability in network services. We propose minor extensions to the packet queueing and discard mechanisms used in routers, coupled with simple control mechanisms at the source that enable the network to guarantee minimal levels of throughput to different sessions while sharing the residual network capacity in a cooperative manner. The service realized by the proposed mechanisms is an interpretation of the controlled-load service being standardized by the Internet Engineering Task Force. Although controlled-load service can be used in conjunction with any transport protocol, our focus in this paper is on understanding its interaction with Transmission Control Protocol (TCP). Specifically, we study the dynamics of TCP traffic in an integrated services network that simultaneously supports both best-effort and controlled-load sessions. In light of this study, we propose and experiment with modifications to TCP's congestion control mechanisms in order to improve its performance in networks where a minimum transmission rate is guaranteed. We then investigate the effect of network transients, such as changes in traffic load and in service levels, on the performance of controlled-load as well as best-effort connections. To capture the evolution of integrated services in the Internet, we also consider situations where only a selective set of routers are capable of providing service differentiation between best-effort and controlled-load traffic. Finally, we show how the service mechanisms proposed here can be embedded within other packet and link scheduling frameworks in a fully evolved integrated services Internet