Integrated Evaluation of Performance and Technology—Throughput of Optical Burst Switching Nodes Under Dynamic Traffic

In this paper, we introduce an integrated analysis of quality-of-service (QoS) performance and technology for optical burst switching (OBS) nodes. Such an integrated approach is needed to comprehensively assess OBS nodes as both burst losses and signal degradation limit their maximum achievable throughput. We apply this analysis to the tune-and-select (TAS) node architecture as well as to extensions with fiber delay line (FDL) buffers and with limited tuning range converters. First, we discuss OBS and motivate the need for the proposed integrated analysis. We survey contention resolution schemes for OBS and discuss principal QoS performance results for wavelength conversion and FDL buffering. Also, we describe the basic and extended TAS architectures as well as their key physical limitations regarding maximum number of wavelengths and maximum throughput. For the different node architectures, we then present the results obtained in the integrated analysis. We compare them based on the maximum utilization as well as the maximum effective throughput under dynamic traffic. Finally, we compare the architectures with respect to their normalized cost per terabit per second (Tb/s) and study its sensitivity with respect to the cost ratio of their key components.      

基于FDL和TWC的光突发竞争解决方案及性能分析

在光突发交换(OBS)网络中,突发竞争是影响网络性能的一个重要因素,因此如何有效地解决它,成了OBS网络非常重要的问题.在分析当前文献中的解决方案的优缺点后,提出了一种突发竞争解决方案的系统实现.该实现将光纤延迟线(FDL)时城缓存与波长转换器(TWC)波长变换、空城技术结合在一起,构造了一个基于前向转发缓存和反馈循环缓存的两级交换结构.最后从多方面对该系统实现的竞争解决有效性进行了性能分析和计算机仿真,结果表明:它在适当的业务强度(小于0.6)下,能有效改善突发丢失率和突发延迟;同时能降低系统所需的光器件数目.     

光突发交换网络新型偏射路由方案研究

光突发交换(OBS)被认为是下一代光网络中的有效核心交换技术之一,设计OBS网络的最初目的之一是减小突发包丢失率.解决突发竞争的方法主要包括光缓存、波长变换、偏射路由和突发分段.提出一种改进偏射路由方案,并建立了数学模型,对改进方案的性能进行了仿真分析,结果表明改进方案能更好的提高网络整体性能.     

An adaptive load-aware burst assembly scheme to achieve optimal performance of FDL buffers in OBS networks

Optical burst switching (OBS) is regarded as one of the most promising switching technologies for next generation optical networks. Contention resolution of data bursts is a critical mission to implement practical OBS. The use of fiber delay line (FDL) buffers has received a lot of attention as a fundamental but effective solution to resolve burst contention. Several studies have investigated the way to achieve the optimal performance of FDL buffers at a single-node level. However, this article studies how to achieve the best performance of OBS networks with FDL buffers under varying traffic condition at a network level. For this purpose, we propose an adaptive load-aware burst assembly (ALBA) scheme, which adaptively adjusts the size threshold of burst assembler optimized to the current network traffic load. A piggybacking method used to deliver the traffic-load information from core nodes to ingress edge nodes accelerates the adaptiveness of the proposed scheme by reducing the update time of the size threshold. The effectiveness of the ALBA scheme is proved by comparing with No-FDL case and fixed size-threshold cases under changing traffic-load environment from extensive simulation tests.

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.     

A burst segmentation-deflection routing contention resolution mechanism in OBS networks

One of the key problems to hinder the realization of optical burst switching(OBS) technology in the core networks is the losses due to the contention among the bursts at the core nodes.Burst segmentation is an effective contention resolution technique used to reduce the number of packets lost due to the burst losses.In our work,a burst segmentation-deflection routing contention resolution mechanism in OBS networks is proposed.When the contention occurs,the bursts are segmented according to the lowest packet loss probability of networks firstly,and then the segmented burst is deflected on the optimum routing.An analytical model is proposed to evaluate the contention resolution mechanism.Simulation results show that high-priority bursts have significantly lower packet loss probability and transmission delay than the low-priority.And the performance of the burst lengths,in which the number of segments per burst distributes geometrically,is more effective than that of the deterministically distributed burst lengths.     

A performance study on optical burst switched networks: the ring topology

Existing performance studies on optical burst switched (OBS) networks have been focusing on channel blocking, i.e., when the required wavelength is not available on a link along a burst’s route. However, we identify another type of blocking as the receiver blocking, i.e., when the receiver at a burst’s destination node is occupied by another burst. Receiver blocking may account for the much larger part of total burst blocking. Receiver blocking has been largely ignored in existing research. In this paper we propose using multiple receivers or fiber delay lines (FDL) in front of the receivers in order to reduce the receiver blocking probability. Extensive simulation results on bi-directional OBS rings are presented to illustrate the problem and the performance of our proposed approaches. The results indicate that receiver blocking can be almost eliminated by using as little as three receivers and can be reduced by using a FDL of small length. To our knowledge, this is the first performance study that covers both channel and receiver blocking for OBS networks.     

A tree-based slot allocation algorithm for loss-free slotted OBS networks

Optical burst switching (OBS) is regarded as one of the most promising switching technologies for next generation optical networks. However, the data burst contention problem is still unresolved thoroughly even though slotted OBS (SOBS) is studied as a new paradigm reducing the blocking rate. In this article, we propose a tree-based slot allocation (TSA) algorithm for loss-free SOBS networks, where the TSA algorithm originally avoids contention of the time-slots by reserving the time-slots with different time-slot positions for the source nodes, respectively. In order to manage the time-slots efficiently, we also propose an OBS superframe, which is a cyclic period and consists of multiple time-slots transmitted by the source nodes toward the same incoming port of a destination node. In addition, we attempt to optimize multiplexing of the OBS superframes to reduce wavelength consumption. On the other hand, when incoming traffic is beyond expectation, a source node may need more time-slots to prevent packet loss because of buffer overflow. For reallocation of the time-slots, we propose a flow control scheme managing some number of shared time-slots, where a control node adaptively allocates (or redeems) the time-slots to (or from) source nodes by utilizing the shared time-slots based on fluctuating traffic condition. Simulation results show that the blocking rate of the proposed TSA–OBS scheme is zero with acceptable queueing delay at moderate traffic offered loads. In addition, multiplexing optimization simulated in the 14-node NSFNET achieves a 63% reduction of wavelength consumption. Moreover, the proposed flow control scheme assisting the TSA algorithm maintains a target upper-bound of queueing delay at the source node, so that packet loss caused by buffer overflow is prevented.     

A novel contention solution strategy based on priority for optical burst switching networks

A fundamental issue in optical burst switching (OBS) networks is to solve the burst contention for the core node. In this paper, a novel priority-based contention solution strategy for OBS networks is proposed. When the contention occurs, the burst priority is considered firstly, and then the burst segmentation method is used for the low priority bursts in this strategy. Ensuring the integrity of high priority bursts, part of the segmented bursts can be transmitted to the destination node via combining wavelength conversion and optical buffer method. Simulation results show that the proposed scheme not only ensures the integrity of high priority bursts, but also reduces the packet loss rate of the low priority bursts maximally, so that it can support good quality of service (QoS) for the network.     

定长光突发下的FDL缓存和调度

为了改善光突发交换的阻塞性能,分析了定长光突发交换在FDL光缓存的排队和调度,提出了基于中间共享FDL光缓存结构的长队列优先(LQP)调度方案,该方案使交换端口充分共享FDL缓存单元,调度时间短,在模拟仿真中得到较好的阻塞性能结果。     

Contention resolution and QoS provisioning using sparse fiber delay lines in WDM networks

In this paper, we combine fiber delay lines (FDL) and optical wavelength conversion (OWC) as the solution for the burst contention problem in optical burst switching (OBS). We present a placement algorithm, k-WDS, for the sparse placement of FDLs at a set of selected nodes in the network. The algorithm can handle both uniform and non-uniform traffic patterns. Our extensive performance tests show that k-WDS provides more efficient placement of optical fiber delay lines than the well-known approach of placing the resources at nodes with the highest experienced burst loss. Performance results are also given to compare the benefit of using FDLs alone, OWCs alone, as well as a mixture of both FDLs and OWCs. A new algorithm, A-WDS, for the placement of an arbitrary numbers of FDLs and OWCs is presented and evaluated under different uniform and non-uniform traffic loads using network simulation of the NSFNET topology and randomly generated graphs. The paper is concluded by presenting the design of a cost-effective optical switch equipped with variable-delay FDL bank. Based on the switch design, a scheme to provide differentiated services for multiple classes of traffic is presented and evaluated.     

OBS网络中基于优先级的先分割后缓存冲突解决方法

为了有效地降低突发包的丢失率和保证OBS网络中不同优先级业务的服务质量,提出了一种基于优先级的先分割后缓存冲突解决方法。该方法通过在多波长信道系统的输出数据信道上设置光纤延迟线组来缓存被分割的冲突突发包。当冲突发生时,首先基于突发包的优先级和处理完毕时间进行"竞争突发包头部分割或者原突发包尾部分割"处理;无冲突部分进行交换或者直接在事先预留的输出数据信道上处理,冲突部分的分割突发包进行光缓存。仿真结果表明,多波长信道系统中,高优先级突发包的丢失率低于低优先级突发包的丢失率。同时该方法在一定程度上可以有效地减少端到端的传输时延和整个网络的丢包率,从而提高整个OBS网络的性能。     

Maximum Size and Throughput of SOA-Based Optical Burst Switching Nodes With Limited Tuning-Range Wavelength Converters and FDL Buffers

In this paper, we show the physical limitations of semiconductor optical amplifier (SOA)-based switching nodes for optical burst switching (OBS). The investigation covers one basic tune-and-select (TAS) architecture, as well as extensions with wavelength converters of limited tuning range (LTR) and dedicated and shared fiber delay line (FDL) buffers. Their maximum size and throughput are analyzed for nodes with four input/output fibers at different bit rates. As main impairments for signal degradation amplifier noise and crosstalk are considered. It is shown that the advanced contention resolution functionality leads to smaller-sized nodes due to higher losses and noise. Among the considered node architectures the highest maximum throughput can be achieved with TAS-LTR due to the smaller number of SOA noise contributions.      

Wavelength Selection in OBS Networks Using Traffic Engineering and Priority-Based Concepts

A fundamental assumption underlying most studies of optical burst switched (OBS) networks is that full wavelength conversion is available throughout the network. In practice, however, economic and technical considerations are likely to dictate a more limited and sparse deployment of wavelength converters in the optical network. Therefore, we expect wavelength assignment policies to be an important component of OBS networks. In this paper, we explain why wavelength selection schemes developed for wavelength routed (circuit-switched) networks are not appropriate for OBS. We then develop a suite of adaptive and nonadaptive policies for OBS switches. We also apply traffic engineering techniques to reduce wavelength contention through traffic isolation. Our performance study indicates that, in the absence of full conversion capabilities, intelligent choices in assigning wavelengths to bursts at the source can have a profound effect on the burst drop probability in an OBS network.     

一种配置有固定波长变换器的OPS节点性能分析

提出一种新型的光分组交换（OPS）节点结构,即OFBFWC（Optical packet switch with Feedback Buffer and Fixed Wavelength Converter）交换结构,它使用固定波长变换器和反馈式的光缓存来解决光分组的冲突。为了提高OFBFWC的性能,还提出一种有效的控制算法。最后,使用仿真实验的方法对OFBFWC的性能进行评估。     

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.     

光突发交换网络中冲突解决方法研究

光突发交换（OBS）网络被认为是下一代光因特网的典型代表,其中,光突发数据包在节点的竞争问题是OBS网络中需要解决的关键问题之一。该文重点研究了突发数据包之间因争夺链路资源而导致冲突的四种竞争解决方案：波长转换、光缓存、偏射路由和突发包分割,并分析了它们的优缺点。在此基础上,提出了OBS网络中冲突解决方法的发展趋势——...     

采用动态波长分插复用器的光突发交换环网中的控制协议研究

提出了一种光突发交换(OBS)环网控制协议——延迟光突发固定周期(DBFP)协议。它采用动态波长分插复用器(WADM)。即采用波长可调发送可调接收(TTTR)机制．通过提前的光突发控制分组(BCP)．为光突发数据分组(BDP)预留资源，并利用固定长度的光纤延迟线(FDL)将突发数据包延迟同定的时间，以便等待所有已经预留的突发数据包发送完毕，从而完全避免突发数据包冲突问题。仿真结果表明。延迟光突发固定周期协议能完全避免突发数据包冲突。同时有效地实现了波长统计复用、按需分配和空间重用，波长重用效率高达160％。从而提高链路利用率．特别适合于突发性的业务；并且延迟光突发固定周期协议所引入的时延为1ms左右．对业务的影响不大。     

Node dimensioning in optical burst switching networks

Network dimensioning should be progressed for pursuing the ultimate efficiency of network system resources in order to satisfy target performance. This article studies node dimensioning as a method of resource optimization in optical burst switching (OBS) networks. OBS is a new switching technology for pursuing bufferless transparent optical networks by sending control packets prior to data burst in order to provision resources for the burst. However, the basic assumption of a bufferless node implies burst contention at a core node when more than two bursts attempt to move forward the same output simultaneously. Thus, burst contention is a critical performance metric and this article takes it into account as a constraint on node dimensioning and target performance. In this article, we first present node dimensioning issues for OBS networks. Two constraints from the transport plane and the control plane which affect burst contention are then introduced. The effect of the burst assembly process on node dimensioning is also presented. From numerical analysis, the optimal number of wavelengths in a link, which provides the lowest blocking probability, is obtained to suggest a guideline for node dimensioning.     

基于偏转路由与COCP的OBS网络竞争解决机制

主要研究解决光突发交换(OBS)网中突发数据包之间争夺链路资源问题的竞争解决机制.从时间域、波长域和空间域上探讨了解决冲突的方法,并给出了一种采用偏转路由机制和基于共享功能波长转换器的冲突部分转换方法(COCP)相结合的竞争解决方法.