Performance modeling of bufferless WDM packet switching networks with limited-range wavelength conversion

All optical communication is attracting more and more attention because of the huge bandwidth of optics. In this paper, we study the performance of bufferless optical wavelength-division multiplexing (WDM) packet switching networks with limited-range wavelength conversion capabilities. We first introduce an optimal scheduling algorithm that maximizes the throughput of the switch. We then derive an analytical model to evaluate the performance of the switch in terms of packet-loss probability. Our model is the first accurate analytical model for a bufferless WDM packet switch with variable conversion distances, and can be used to quantitatively determine the maximum load for a given conversion distance or the minimum conversion distance for a given traffic load. We also conducted simulations to validate the analytical model. Both the analytical and simulation results reveal that limited-range wavelength conversion can achieve almost the same performance as full-range wavelength conversion.     

Dynamically concatenated wavelength converters

An architecture is proposed for a wavelength-division-multiplexed (WDM) optical packet or circuit switch in which a bank of limited-range wavelength converters is shared among all input fibers, and in which any subset of converters can be dynamically concatenated (cascaded) to yield a wider conversion range for a packet that would otherwise be dropped because all unused wavelengths in its desired output fiber lie outside the range of a single converter. A probabilistic model of a switch is used to numerically determine the improvement in packet-drop probability achieved by dynamically concatenating converters.     

Operation of WDM networks with different wavelength conversioncapabilities

We study the impact of wavelength conversion capability on wavelength routing WDM networks with fixed shortest-path routing. We propose a method for implementing wavelength routing in a WDM network with partial wavelength conversion capability. Simulation results show that such partial wavelength conversion networks provide a performance in between that of wavelength continuous networks and those with full conversion capability. In addition, it can be seen that only limited wavelength conversion capability is enough to provide a performance close to that of a network with full conversion. Analytical and simulation bounding results for the full and no conversion cases have also been provided     

Analytical MMAP-Based Bounds for Packet Loss in Optical Packet Switching with Recirculating FDL Buffers

The major goal of optical packet switching (OPS) is to match switching technology to the huge capacities provided by (D)WDM. A crucial issue in packet switched networks is the avoidance of packet losses stemming from contention. In OPS, contention can be solved using a combination of exploitation of the wavelength domain (through wavelength conversion) and buffering. To provide optical buffering, fiber delay lines (FDLs) are used. In this paper, we focus on an optical packet switch with recirculating FDL buffers and wavelength converters. We introduce the Markovian arrival process with marked transitions (MMAP), which has very desirable properties as a traffic model for OPS performance assessment. Using this model, we determine lower and upper bounds for the packet loss rate (PLR) achieved by the aforementioned switch. The calculation of the PLR bounds through matrix analytical methods is repeated for a wide range of traffic conditions, including highly non-uniform traffic, both in space (i.e., packet destinations) and time (bursty traffic). The quality of these bounds is verified through comparison with simulation results.     

Dimensioning of the Wavelength Converters in a WDM Optical Packet Switch

This paper analyzes the performances of a WDM optical packet switch making use of fiber delay lines to resolve output packet contentions; the optical packet switch is equipped with tunable wavelength converters which can shift optical packets to any wavelength of the output link which they are directed to. As not all packets need conversion, we propose a dimensioning technique allowing to reduce the number of converters and to improve the signal quality by reducing unnecessary conversions. The obtained results show that a remarkable reduction of the number of converters is obtained with respect to that needed by other switches described in literature. Such a saving is achieved by maintaining the packet loss probability below a prefixed threshold.     

Analytical model of sparse-partial wavelength conversion in wavelength-routed WDM networks

Wavelength conversion is one of the key techniques to improve the blocking performance in wavelength-routed WDM networks. Given that wavelength converters nowadays remain very expensive, how to make effective use of wavelength converters becomes an important issue. In this letter, we analyze the sparse-partial wavelength conversion network architecture and demonstrate that it can significantly save the number of wavelength converters, yet achieving excellent blocking performance. Theoretical and simulation results indicate that, the performance of a wavelength-routed WDM network with only 1-5% off wavelength conversion capability is very close to that with full-complete wavelength conversion capability.     

Multi-wavelength switching in IP optical nodes adopting different buffering strategies

The paper addresses the topic of long-haul optical networking for the provision of large-bandwidth IP services. A class of optical packet switching architectures is considered which adopts an arrayed wavelength grating device as packet router. The architecture performs slotted packet switching operations and fully exploits the wavelength routing capabilities by allowing multi-wavelength switching. Fiber delay lines are used to perform optical packet buffering, which accomplishes either input queueing or shared queueing. Here a thorough performance evaluation is carried out with different buffering configurations and the effect of various switch parameters on traffic performance is studied.     

Design of grooming architectures for optical WDM mesh networks: limited grooming with electronic wavelength conversion

In this article, we consider the problem of traffic grooming in optical wavelength division multiplexed (WDM) mesh networks under static traffic conditions. The objective of this work is to minimize the network cost and in particular, the electronic port costs incurred for meeting a given performance objective. In earlier work, we have shown the benefits of limited grooming switch architectures, where only a subset of wavelengths in a network are equipped with expensive SONET Add Drop Multiplexers (SADM) that provide the grooming functionality. In this work, we also consider the wavelength conversion capability of such groomers. This can be achieved using a digital cross-connect (DCS) in the grooming switch to switch low-speed connections between the SADMs (and hence, between wavelengths). The grooming switch thus avoids the need for expensive optical wavelength converters. Based on these observations, we propose a limited conversion-based grooming architecture for optical WDM mesh networks. The local ports at every node in this architecture can be one of three types: an add-drop port, a grooming port that allows wavelength conversion or a grooming port that does not allow wavelength conversion. The problem studied is: given a static traffic model, where should the different ports be placed in a network? We formulate this as an optimization problem using an Integer Linear Programing (ILP) and present numerical results for the same. We also present a heuristic-based approach to solve the problem for larger networks.     

All-Optical Wavelength-Bypassing Ring Communications Networks

We introduce an all-optical WDM packet communication network that performs wavelength bypassing at the routers. Packets that arrive at a wavelength (optical cross-connect) router at designated wavelengths are switched by the router without having their headers examined. Thus, the processing element of the router is bypassed by such packets. For packet traffic that uses wavelengths that do not bypass a switch, the headers of such packets are examined to determine if this switch is the destination for the flow. If latter is the case, the packet is removed. Otherwise, the packet is switched to a pre-determined output without incurring (network internal) queueing delays. We study a ring network with routers that employ such a WDM bypassing scheme. We present methods to construct wavelength graphs that define the bypassing pattern employed by the routers to guide the traffic flows distributed at each given wavelength. Performance is measured in terms of the network throughput and the average processing path length (i.e., the average number of switches not being bypassed). For a fixed total processing capacity, we show that a WDM bypassing ring network provides a higher throughput level than that exhibited by a non-bypassing ring network, using the same value of total link capacity. By using WDM bypassing, the average processing path length (and thus the packet latency) is reduced. We study a multitude of network loading configurations, corresponding to distinct traffic matrices and client-server scenarios. Higher throughput levels are obtained for network configurations driven by non-uniform traffic matrices. The demonstrated advantages of WDM bypassing methods shown here for WDM ring networks are also applicable to more general network topological layouts.     

Architectures and Performance of Optical Packet Switching over WDM

Packet switching over wavelength division multiplexing (WDM) channels is considered with the aim to investigate algorithms for wavelength assignment and to define feasible switch architectures, in the presence of connectionless or connection-oriented transfer modes. In particular, as regards the connection-oriented scenario, mapping of virtual connections onto wavelengths operated by network nodes is considered and procedures are proposed to achieve statistical multiplexing efficiency by dynamic wavelength re-assignment. Switch architectures to support dynamic wavelength encoding and the related performance evaluation are presented and discussed in the paper, evidencing the benefits of packet switching over WDM.     

Cost Evaluation of Optical Packet Switches Equipped With Limited-Range and Full-Range Converters for Contention Resolution

Two architectures are proposed for a wavelength-division multiplexed optical packet switch equipped with both limited-range wavelength converters (LRWCs) and shared full-range wavelength converters (FRWCs). The FRWCs are used to overcome the performance degradation in terms of packet loss probability due to the use of LRWCs only. Two different sharing strategies of the FRWCs are considered. In the first architecture, a pool of FRWCs is shared among the arriving packets. In the second one, the sharing is only partial and the packets directed to the same output share a same pool of FRWCs. A probabilistic model is proposed to dimension the number of shared FRWCs so that the same packet loss probability of a switch equipped with only shared FRWCs is guaranteed. After introducing a cost model of the converters depending on the conversion range, we show that the architectures may allow a conversion cost savings on the order of 90%.     

波长路由WDM网络中"稀疏-部分"波长转换的建模与分析

分析稀疏-部分波长转换网络的结构,论证了该结构在获得优秀的阻塞性能同时,能显著地节省波长转换器的数目.     

WDM packet routing for high-capacity data networks

We present experimental and numerical studies of a novel packet-switch architecture, the data vortex, designed for large-scale photonic interconnections. The selfrouting multihop packet switch efficiently scales to large port counts (>10 k) while maintaining low latencies, a narrow latency distribution, and high throughput. To facilitate optical implementation, the data-vortex architecture employs a novel hierarchical topology, traffic control, and synchronous timing that act to reduce the necessary routing logic operations and buffering. As a result of this architecture, all routing decisions for the data packets are based on a single logic operation at each node. The routing is further simplified by the employment of wavelength division multiplexing (WDM)-encoded header bits, which enable packet-header processing by simple wavelength filtering. The packet payload remains in the optical domain as it propagates through the data-vortex switch fabric, exploiting the transparency and high bandwidths achievable in fiber optic transmission. In this paper, we discuss numerical simulations of the data-vortex performance and report results from an experimental investigation of multihop WDM packet routing in a recirculating test bed     

A Scalable and High Capacity All-Optical Packet Switch: Design,Analysis, and Control

In this paper, a modular and scalable all-optical packet switch (AOPS) is proposed. The range of its capacity can be easily scaled from gigabit per second to multi-terabits per second. Due to its broadcast-and-select property, the proposed AOPS is capable of performing a multicast function. By taking the advantage of wavelength division multiplexing (WDM), this architecture can provide the best network performance using a limited number of optical fiber delay lines as buffers. To perform the header replacement function, a novel all-optical header replacement unit (HRU) is introduced to be integrated with the switching function. The proposed HRU is shared by all the inputs which provides cost advantages. In addition, we present a generic control scheme for the proposed AOPS. To implement the function of the AOPS, two possible approaches, based on the design of wavelength conversion pools (WCPs), are presented and their cascadability performances are compared. Our simulations show that the proposed AOPS with an arrayed waveguide grating (AWG) based WCP provides better cascadability performance than the one with a star coupler based WCP. We conclude that, based on the status of current optical and electronic technologies, the proposed architecture is feasible to be implemented, and can be a good candidate for future packet switching solutions.     

一种新型的输出缓存式光分组交换节点性能分析

光缓存和波长变换是解决光分组竞争的有效方法.提出一种称为FCOB的新型光分组交换节点结构,它使用固定波长变换器和输出式光缓存来解决光分组的竞争.针对FCOB交换结构,还提出一种有效的竞争控制算法.最后,使用仿真实验对FCOB的性能进行了评估.仿真结果表明,虽然使用的是固定波长变换器,但FCOB交换结构仍具有良好的性能.     

Performance Analysis of Multi-Fiber WDM Network with Limited-Range Wavelength Conversion

Wavelength routed optical networks have emerged as a technology that can effectively utilize the enormous bandwidth of the optical fiber. Wavelength conversion technology and wavelength converters play an important role in enhancing fiber utilization and in reducing the overall call blocking probability of the network. In this paper, we develop a new analytical model to calculate the average blocking probability in multi-fiber link networks using limited range wavelength conversion. Based on the results obtained, we conclude that the proposed analytical model is simple and yet can effectively analyze the impact of wavelength conversion ranges and number of fibers on network performance. Also a new heuristic approach for placement of wavelength converters to reduce blocking probabilities is explored. Finally, we analyze network performance with the proposed scheme. It can be observed from numerical simulations that limited range converters placed at a few nodes can provide almost the same blocking probability as full range wavelength converters placed at all the nodes. We also show that being equipped with a multi-fiber per-link has the same effect as being equipped with the capability of limited range wavelength conversion. So a multi-fiber per-link network using limited range wavelength conversion has similar blocking performance as a full wavelength convertible network. Since a multi-fiber network using limited range wavelength conversion could use fewer converters than a single-fiber network using limited range wavelength conversion and because wavelength converters are today more expensive than fiber equipment, a multi-fiber network in condition with limited range wavelength conversion is less costly than a single fiber network using only limited range wavelength conversion. Thus, multi-fiber per-link network using limited range wavelength conversion is currently a more practical method for all optical WDM networks. Simulation studies carried out on a 14-node NSFNET, a 10-node CERNET (China Education and Research Network), and a 9-node regular mesh network validate the analysis.     

Wavelength conversion and call connection probability in WDMnetworks

Call connection probability in an all-optical network using wavelength division multiplexing (WDM) depends on the number of WDM wavelengths and on the network capability for wavelength conversion at network nodes. New closed-form expressions for call connection probability are derived which reflect these dependencies. These expressions are combined in a converter availability model to study the impact of wavelength conversion on call connection probability and the tradeoff between network wavelength conversion capability and the number of WDM wavelengths     

基于共享有限波长和参量波长转换器的冲突解决方案

研究了一种共享有限波长转换器(LRWC)和参量波长转换器(PWC)相结合的解决全光分组波长资源竞争的方案,并提出了PWC优先算法作为光分组交换中的波长竞争解决方案。仿真结果表明:本文结构与共享LRWC节点结构相比,减少了波长转换器的数目,并且能够明显降低光分组交换节点的分组丢包率(PLP),提高波长转换器的利用率。研究还发现,不同波长转换器配置方案对系统性能有影响。     

波分复用波长路由节点的阻塞特性分析

利用概率统计理论的方法，从节点层次上定量分析了节点规模、复用波长数目以及波长转换对波分复用(WDM)波长路由网络中波长路由节点的影响。提出了基于概率统计的节点阻塞模型。数值结果突出表明波长转换能力越强的全光节点，其性能越优。为了提高网络资源的使用效率并增强全光网络的灵活性，必须实现全光网络中的虚波长路由波长转换器。通过数值计算找到了阻塞性能和代价的折中，研究中发现配置较低波长转换能力波长转换器的波长路由节点将会具备更强的性价比优势，当前在构建光通信系统时使用弱波长转换能力的光节点更可行。     

Scheduling Algorithms for a Slotted Packet Switch with either Fixed or Variable Length Packets

We address the problem of congestion resolution in optical packet switching (OPS). We consider a fairly generic all-optical packet switch architecture with a feedback optical buffer constituted of fiber delay lines (FDL). Two alternatives of switching granularity are addressed for a switch operating in a slotted transfer mode: switching at the slot level (i.e., fixed length packets of a single slot) or at the burst level (variable length packets that are integer multiples of the slot length). For both cases, we show that in spite of the limited queuing resources, acceptable performance in terms of packet loss can be achieved for reasonable hardware resources with an appropriate design of the time/wavelength scheduling algorithms. Depending on the switching units (slots or bursts), an adapted scheduling algorithm needs to be deployed to exploit the bandwidth and buffer resources most efficiently.