Number Of Fiber Delay Lines Research Articles

Optical Packet and Burst Switched Networks Performance for Time-Dependent Video Traffic

In this paper we study the performance of optical packet and burst switched networks (OSN) to supply video traffic with scaling characteristics. The optical buffering generally implemented by fiber delay lines (FDLs) is the main challenge to achieve a full-optical switching architecture. For the analysis the MIGIoo and fBm processes are considered to represent the time-dependent video traffic. These stochastic processes are examined in terms of its video traffic scaling (H), heavy-tail distribution (HTD) and autocorrelation (ACF) characterization. We show that with a small number of FDLs (25-50) of 33.92m each and no wavelength conversion an optical packet switch can achieve performance results similar to a burst architecture. Moreover, the jitter values introduced by these FDLs were lower compared to the jitter requirements of the more demanding video traffic. We also demonstrate that the video traffic scaling degree was not affect by the optical packet or burst buffeting solutions. However, the video traffic tail distribution showed to be the critical aspect to the optical switches dimensioning and thus the future optical switched network performance.

An Optical Packet Switch with Recirculation Limited Range Wavelength Converter Groups and Recirculation Optical Buffers

AbstractThis paper focuses on performances of limited range wavelength converters (LRWCs) in resolving optical packet contentions. An optical packet switch is proposed, which uses recirculation LRWC groups and recirculation fiber delay lines (FDLs) to resolve optical packet contentions. In order to make full use of the LRWCs and the FDLs, a control algorithm is also proposed. Packet loss probability of the proposed switch versus the number of FDLs, the wavelength conversion range, and the number of LRWC groups are evaluated in detail.

WDM Recirculation Buffer-Based Optical Fabric for Scalable Cloud Computing

Optical packet switching is the next-generation disruptive technology for massive cloud data centers. All-optical routers, supporting wavelength parallelism, accelerate the execution of applications and facilitate server virtualization. Despite the recent advances in optical technologies, the greatest challenge in the realization of optical routers is the lack of a mature solution for buffering optical signals. Fiber delay lines (FDLs) have been proposed to emulate electronic buffering by delaying contending optical packets for a fixed amount of time. However, the practical limitation on the number of FDLs in a router requires its ports to be run at low utilization, sacrificing a significant portion of network capacity. In this paper, we introduce a novel modular router architecture based on wavelength-division multiplexed (WDM) recirculation buffers as a solution to the problem of limited buffering capacity. We propose a load-balancing scheduler to maximize throughput. Our mathematical analysis and Monte Carlo simulations show that the consolidation of optical buffers in WDM FDLs accompanied with internal load balancing leads to a virtually lossless router, resilient to data center traffic anomalies. The load balancing further minimizes the number of packet recirculations, leading to negligible queueing latency. Our design supports physical layer scalability and is amenable to large-scale optical integration.

An irregularly slotted ring scheme for contention-free optical burst switching

Optical burst switching (OBS) is a promising way to efficiently use the massive bandwidth in wavelength division multiplexing (WDM) networks. However, the problem of contention between bursts is hard to be solved in OBS networks with general topology. For that reason, the conception of contention-free OBS ring is proposed. In recent years, several OBS ring schemes have been designed. According to the number of transceivers installed at each node, the schemes can be roughly divided into two classes: single-transceiver schemes and multi-transceivers schemes. Among which, multi-transceivers schemes are more efficient than single-transceiver ones. There are two existing multi-transceivers schemes, LightRing and SWING, yet both of them have disadvantages. The LightRing scheme has a high queuing delay and needs large buffers, while the SWING scheme not only needs a large number of fiber delay lines but also has a specific requirement to the girth of the ring. In this paper, we design another multi-transceivers OBS ring scheme named irregularly slotted OBS ring (IS-OBS-Ring). In the IS-OBS-Ring scheme, all data wavelengths are slotted and slots on different wavelengths are not synchronously but irregularly distributed. Simulation results show that the performance of IS-OBS-Ring is further better than the LightRing scheme and similar to the SWING scheme. However, the IS-OBS-Ring scheme can overcome the disadvantages of both the LightRing scheme and the SWING scheme.

Efficient Designs of Optical LIFO Buffer with Switches and Fiber Delay Lines

The lack of optical buffer is still one of the main problems that hinder the development of all optical networks. Current works on this topic mainly focus on the emulation of optical buffers based on a combination of fiber delay line (FDL) and switch. These works have demonstrated the feasibility of FDL-based emulation for many kinds of optical buffers, like the First In First Out (FIFO) buffer, Priority buffer, etc. The Last In First Out (LIFO) buffer is another basic network component for congestion control and QoS guarantee. Recently, Huang et al. introduced a recursive construction for the LIFO buffer, which requires no less than 9 log2 B FDLs to build such a buffer of size B. In this paper, we first show that by a proper FDL grouping and a suitable FDL length assignment for each FDL-group, we need approximately 3 log2 B FDLs to emulate a LIFO buffer of size B. We then demonstrate that if a careful packet scheduling among FDL-groups is adopted, this number of FDLs can be further reduced to 2 log2 B.

Average Number of Recirculations in SDL Constructions of Optical Priority Queues

In this letter, we derive the average number of times an optical packet recirculates through the optical switch and the fiber delay lines in our previous constructions of optical priority queues (see Figure 1 in Section I) under Bernoulli arrival traffic, Bernoulli control input, and uniform priority assignment. The analytical results on the average number of recirculations are further verified through simulations. Through simulations, we also find that these analytical results are useful in choosing the number of fiber delay lines in our constructions of optical priority queues when there is a limitation on the number of times an optical packet can be recirculated through the optical switch and the fiber delay lines.

Performance analysis of an improved burst-outputted scheme in a limited buffer equipped OBS core node

Burst-outputted performance will be affected by the outputting scheme in optical burst switching (OBS) core nodes. In this paper, we have evaluated the performance of an improved burst-outputted scheme in an OBS core node, in which fiber delay lines (FDLs) based limited optical buffer has been equipped. Different from the previous works which only consider the same probability to select output-port for burst, the proposed scheme gives a burst-outputted process based on different port selected probability. Simulation results have indicated that, performance evaluation parameters are changed with the output-port selected probability and the different number of FDLs in an optical buffer.

A framework for fiber delay-line buffers in packet-based asynchronous multifiber optical networks (PAMFONET)

A major challenge in packet-based optical networks is packet contention, which occurs when two or more packets are heading to the same output at the same time. To resolve contention in the optical domain, a fundamental approach is fiber delay-line (FDL) buffering, in which packets can be delayed for a fixed amount of time. In the literature, the performance of FDL buffering has been studied extensively. However, most existing works are based on an assumption that there is only one fiber per link in the network. In this paper, we address the architecture and performance of FDL buffers in packet-based asynchronous multifiber optical networks (PAMFONET), in which each link in the network may consist of multiple optical fibers. We propose a framework for FDL buffers in PAMFONET, in which we provide three essential architectures and corresponding packet scheduling policies. Extensive simulation results show that, with appropriate settings, the same number of FDLs can lead to better performance in multifiber networks than in single-fiber networks. Copyright © 2011 John Wiley & Sons, Ltd. (Here we note that such kind of multistage architectures are widely used in design and implementation of FDL buffers [14].

Two-stage shared fiber delay line optical packet switch for contention resolution in optical packet-switching network

Packet contention is a major issue in an optical packet-switching network. We focus on contention resolution using fiber delay lines (FDLs), which are used to buffer contended packets in order to forward these packets to the desired output port to some future time slots. Most of the existing optical buffering schemes are output based, which require a huge number of FDLs as well as a larger switch size and incur extra implementation cost. A shared FDL buffering approach is considered to reduce implementation cost. This paper proposes a two-stage shared FDL optical packet switch to resolve contention, where FDLs are implemented in two stages using extremely simple auxiliary switch. The performance of the proposed switch architecture is evaluated through simulation under trivial and bursty traffic condition, in terms of packet loss rate and average delay. Other performance parameters, such as cross-talk, insertion loss, switch time, etc., are also discussed. Simulation result shows that, the packet loss rate of the proposed architecture under heavy traffic load ( = 0.9) is between ~10−3 and ~10−4 for a 32×32 switch using 64 FDLs, and zero packet loss rate achievable with < 0.8.

Performance of two-stage shared fiber delay line optical packet switch under bursty traffic

Packet contention is a major issue in an optical packet switching network. It is not a trivial task to resolve contention due to lack of optical RAM technology. This article proposes a two-stage shared fiber delay line (FDL) optical packet switch for contention resolution. In this article, shared FDLs are used to buffer optical packets, in which a pool of buffer memory is shared among all switch output ports. Most of the existing optical buffering schemes are output-based which require a huge number of FDLs as well as a larger switch size that incur extra implementation cost. However, a shared buffering approach is considered in this article in order to reduce implementation cost. In this article, FDLs are implemented in two stages using an extremely simple auxiliary switch. The proposed switch architecture leads to more efficient use of buffer space. The superiority of the proposed switch architecture has been established by means of extensive simulations. The performance of the proposed switch is investigated under bursty traffic. Simulation result shows that the proposed switch can achieve satisfactory performance at the price of a reasonable amount of FDLs. Moreover, the significance of the proposed switch is confirmed by simulation.

An effective buffering architecture for optical packet switching networks

A novel optical buffering architecture for Optical Packet Switching (OPS) networks is proposed in this article. The architecture which adopts a fiber-sharing mechanism aims at solving the problem of using a large number of fiber delay lines that are used to solve resource contention in the core node in OPS networks. The new architecture employs fewer fiber delay lines compared to other simple architectures, but can achieve the same performance. Simulation results and analysis show that the new architecture can decrease packet loss probability effectively and achieve reasonable performance in average packet delay.

Feasibility analysis for service differentiation using an FDL bank in OBS networks

A service differentiation scheme in optical burst switching (OBS) networks, which is based on dynamic fiber delay line (FDL) assignment, is shown. The effectiveness of the scheme is validated by numerical analysis and extensive simulations. Especially, the feasibility conditions for the service differentiation scheme, which are considered as the minimum number of FDLs for each sub-FDL group, are displayed. The feasibility conditions are derived numerically, and are verified through extensive simulations. The results of extensive simulations show that the proposed scheme and the feasibility conditions are valid for service differentiation in OBS networks.

Service differentiation scheme in OBS networks

In order to achieve service differentiation, especially loss differentiation, in optical burst switching (OBS) networks, we propose a dynamic fiber delay line (FDL) partitioning algorithm, which divides FDLs into several groups over a feed-forward output buffering architecture. In the proposed scheme, a plurality of traffic classes and FDL groups can be considered, and each FDL group is assigned to traffic classes, so that the target loss probabilities of classes are guaranteed. Also, the optimal number of FDLs for each FDL group by the proposed algorithm is decided in Poisson traffic environments. The extensive simulation results validate the effectiveness of the proposed dynamic FDL partitioning algorithm for the loss differentiation in OBS networks.

A redundant overbooking reservation algorithm for OBS/OPS networks

Optical burst switch (OBS) has been a hot topic in the area of optical networks. OBS can realize statistic multiplexing while having the advantage of simplicity in deployment. A key issue in OBS is how to reduce the rate of the burst dropping caused by the absence of optical buffers or by the limited number of Fiber Delay Line (FDL). Existing work addresses this issue in the wavelength, time, or space domains. This paper proposes an algorithm, which works from a new domain to address this problem. Specially, it works by using the idle bandwidth in upstream links to transmit redundant copies of bursts to provide more sending chances at the downstream node. Simulation results are given to demonstrate the high performance of our algorithm.

Modeling and design of WDM optical buffers in asynchronous and variable-length optical packet switches

This paper aims to analyze the influence of wavelength conversion on the design of optical fiber delay line (FDL) buffers in a wavelength-division multiplexing (WDM) packet switch. We focus on the network scenario that variable-length packets arrive at the optical switch asynchronously. Through a detailed analysis of the behaviors of packets in the WDM FDL buffer, an approximate analytical traffic model is established. And the theoretical model is proved to be reasonable by simulations. Based on the model, it is clear that, under the same traffic load per wavelength channel, the utilization of tunable wavelength converters (TWCs) to resolve the output contention significantly reduces the number of fiber delay lines (FDLs) in optical buffers, and to achieve a given packet loss probability under a certain number of FDLs, the required number of tunable wavelengths can be predicted. The model can also estimate the optimal value of the basic time unit of the FDLs, and examine the influence of WDM on the optimal value. Finally, the impact of wavelength conversion on the optical switch size is analyzed. To the best of our knowledge, it is the first time that modeling on this problem is presented.

Performance evaluation of multi-fiber optical packet switches

Multi-fiber WDM networks are becoming the major telecommunication platforms for transmitting exponentially increasing data traffic. While today's networks are mainly providing circuit-switched connections, optical packet-switching technologies have been investigated for years, aiming at achieving more efficient utilizations of network resources. In this paper, we have evaluated, for the first time, the packet-loss performance of multi-fiber optical packet switches (MOPS). Our main contributions are threefold. Firstly, we have proposed simple and accurate analytical models for analyzing packet-loss performance of (i) the most fundamental MOPS configuration, (ii) MOPS equipped with fiber delay lines (FDLs) and (iii) shared wavelength converters (SWCs). Secondly, we have shown that the MOPS network cannot achieve the same performance as the one with full wavelength conversion (FWC), which is quite different from the well-known conclusion in circuit-switched networks. However, MOPS does significantly outperform the classic single-fiber switches. By introducing a small number of FDLs or SWCs, it outperforms the highly expensive FWC solution as well. Finally, we have taken the hardware constraints into consideration by evaluating the performance of MOPS configurations having multiple limited-sized switching boards, which leads to some insights helpful for developing cost-effective MOPS configurations in the future.

WDM optical interconnects with recirculating buffering and limited range wavelength conversion

All-optical communication, in particular, wavelength-division-multiplexing (WDM) technique, has been proposed as a promising candidate to meet the ever-increasing demands on bandwidth from emerging bandwidth-intensive computing/networking applications. However, with current technology, the cost of optical communication, especially the cost of optical buffering and wavelength conversion, remains a major concern for such applications. In this paper, we study WDM optical interconnects that utilize low cost recirculating buffering and limited range wavelength conversion. We first consider the packet scheduling problem in this type of interconnect, and formalize the problem of maximizing throughput and minimizing packet delay as a matching problem in a bipartite graph. We give an optimal parallel algorithm for this problem that runs in O(Bk/sup 2/) time, compared to O((N+B)/sup 3/k/sup 3/) time if directly applied to existing matching algorithms for general bipartite graphs, where N is the number of input/output fibers of the interconnect, B is the number of fiber delay lines, and k is the number of wavelengths. We also consider efficient switching fabric designs for this type of interconnect. We distinguish between the switching fabric connecting the input fibers to the output fibers and the switching fabric connecting the input fibers to the delay lines and show that by adopting the idea of concentration, the cost of the latter can be reduced significantly in terms of the number of crosspoints.

Burst loss and delay in optical buffers with offset-time management

In this paper, we study the loss and delay of data bursts in an optical buffer. We assume that this buffer consists of a number of fiber delay lines (FDLs). In order to guarantee quality of service (QoS) differentiation in such a buffer, we investigate analytically an offset-time based scheduling mechanism. We consider a system with C QoS classes, where the high-priority QoS classes have a larger offset time than the low-priority QoS classes. For this system, we calculate the total burst loss probability and the burst loss probability within each QoS class. Furthermore, we study the delay of an arbitrary arriving data burst, as well as the delay of an arriving data burst of a certain QoS class.

Ant-based contention resolution in slotted WDM optical packet switched networks

This paper proposed an Ant-based contention resolution scheme for the slotted optical packets switched (OPS) networks. The proposed contention resolution schemes are shown to be effective to resolve packet contention and to achieve good performance by using minimum number of fiber delay lines (FDL). Moreover, the cell loss rate is improved by 10% to 40% under various traffic loads using the proposed scheme.

Using switched delay lines for exact emulation of FIFO multiplexers with variable length bursts

In the literature, research has been published extensively on how one achieves exact emulation of First In First Out (FIFO) multiplexers for fixed-size cells (or packets) using optical crossbar Switches and fiber Delay Lines (SDL). In this paper, we go a step further and propose a new architecture that achieves exact emulation of FIFO multiplexers for variable length bursts. Our architecture consists of two blocks: a cell-scheduling block and a FIFO multiplexer for fixed-size cells. Both blocks are made of SDI, units. The objective of the cell-scheduling block is to schedule cells in a burst to the right input at the right time so that cells in the same burst depart contiguously from the multiplexer for fixed-size cells. We show that cell scheduling can be done efficiently by keeping track of a single state variable, called the total virtual waiting time in this paper. Moreover, the delay through the cell-scheduling block is bounded above by a constant that only depends on the number of inputs and the maximum number of cells in a burst. Such a delay bound provides a limit on the number of fiber delay lines needed in the cell-scheduling block.