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.

Node pacing for small optical RAM-buffered packet-switching networks

One of the difficulties with optical packet switched (OPS) networks is buffering optical packets in the network. The only available solution that can currently be used for buffering in the optical domain is using long fiber lines called fiber delay lines (FDLs), which have severe limitations. Moreover, the research on optical RAM presently being done is not expected to achieve a large capacity soon. However, the burstiness of Internet traffic causes high packet drop rates and low utilization in very small buffered OPS networks. We therefore propose a new node-based pacing algorithm for decreasing burstiness. We show that by applying some simple pacing at the edge or core backbone nodes, the performance of very small optical RAM buffered core OPS networks with variable-length packets can be notably increased.     

Demonstration of 10 Gbit Ethernet/Optical-Packet Converter for IP Over Optical Packet Switching Network

We developed novel network interfaces, for example 10 Gbit Ethernet to 80 Gbit/s optical-packet (10 GbitE–80 GbitOP) or 80 Gbit/s optical-packet to 10 Gbit Ethernet (80 GbitOP–10 GbitE) converters (collectively called as 10 GbitE/80 GbitOP converters), to connect optical packet switching (OPS) networks with IP technology-based networks. By using newly developed arrayed burst-mode optical packet transmitters/receivers together, the 10 GbitE–80 GbitOP converter at the ingress edge node of the OPS network encapsulates an IP packet into an $80(8lambdatimes 10) {rm Gbit/s}$ dense wavelength division multiplexing (DWDM)-based optical packets and generates an optical label based on a lookup table and the destination addresses of the IP packet. The 80 GbitOP–10 GbitE converter at the egress edge node decapsulates the IP packet from the optical packet and generates a 10 GbitE frame accommodating the IP packet according to a lookup table. By using these network interface devices and OPS system based on multiple optical label processing, we achieved, for the first time, 74-km single-mode fiber transmission, switching, and buffering of $80(8lambdatimes 10) {rm Gbit/s}$ DWDM-based optical packets encapsulating almost 10 Gbit/s IP packets with error-free operation (IP packet loss rate $≪ 10^{-6}$).      

Demonstration of 100 Gbps optical packet switching using header processor based on 48-bit longest prefix matching

Wire-rate packet processing and its energy saving for over 100 Gbps speed of line are major issues to be resolved in optical packet switching (OPS) networks. For that purpose, we newly develop a high-speed, deterministic-latency electronic header processor based on longest prefix matching (LPM) for searching optical packet destination addresses (OP-DAs). This paper reports the successful experimental results of electronic header processing based on LPM search of up to 48 bits and optical switching of 100 Gbps optical packets by the use of the header processor. We demonstrate 48-bit LPM-capable optical packet switching. We also demonstrate IP packet transfer and 32-bit LPM-capable optical packet switching. In the latter demonstration, the 32-bit OP-DA of optical packets is directly copied from the 32-bit destination address of Internet Protocol version 4 (IPv4) packets. This result indicates that OPS networks can be deployed with electronic IP networks by the use of integrated network operation between OPS and IP networks.     

Evaluation of QoS differentiation mechanisms in asynchronous bufferless optical packet-switched networks

Existing quality of service differentiation schemes for today's IP over point-to-point optical WDM networks take advantage of electronic RAM to implement traffic management algorithms in order to isolate the service classes. Since practical optical RAM is not available, these techniques are not suitable for a future all-optical network. Hence, new schemes are needed to support QoS differentiation in optical packet-switched (OPS) networks. In this article we first present an overview of existing QoS differentiation mechanisms suitable for asynchronous bufferless OPS. We then compare the performance of the presented schemes and qualitatively discuss implementation issues, in order to evaluate the mechanisms. In particular, we present an evaluation framework, which quantifies the throughput reduction observed when migrating from a best effort scenario to a service-differentiated scenario. Our study shows that preemption-based schemes have the best performance, but also the highest implementation complexity.     

Demonstration of an agile hybrid IP-optical packet construction mechanism in wavelength routed optical packet switched networks

This paper presents a novel architecture for an ingress edge OPS router and demonstrates a mechanism that maps Internet traffic onto optical packets. The architecture utilises a high-speed reconfigurable hardware platform and a fast tunable laser and also supports the user network interface (UNI) functionality by fully interacting with the physical and higher network layers. The main design issues, including wavelength agility, traffic aggregation based on the class of service and variable length optical packet construction and transmission are also discussed.     

Recursive construction of FIFO optical multiplexers with switched delay lines

In this paper, we develop mathematical theory for recursive construction of first-in first-out (FIFO) optical multiplexers by the combination of (bufferless) crossbar switches and fiber delay lines (SDLs). We show that by cascading multistage SDL units, 2-to-1 multiplexers with a large buffer can be emulated exactly for both the departure process and the loss process from the multiplexer. Such results are extended to the case of n-to-1 multiplexers by introducing a new class of multiplexers, called delayed-loss multiplexers. A delayed-loss multiplexer has the same departure process as an ordinary multiplexer. However, lost packets due to buffer overflow in a multiplexer might be delayed. A key result from our theory is the self-routing n-to-1 multiplexer, where the routing path of a packet through the multistage SDL units can be determined upon its arrival.     

On the performance of different node configurations in multi-fiber optical packet-switched networks

With the development of optical packet-switching (OPS) technologies, multi-fiber OPS networks will play an important role in the future data transmissions. In such networks, instead of constructing some extremely expensive node configurations with strictly non-blocking switching function, a more practical solution is multi-board switches that contain a number of small-sized switching boards. In this article, we have evaluated the performance of several different multi-board switches, based on the following two main objectives: (i) better understanding the effects of different connection schemes between switching boards and optical buffers and (ii) investigating possible schemes for achieving comparable performance to that of the ideal, strictly non-blocking switches. Extensive simulation results have shown that unlike circuit-switched net- works, multi-board OPS cannot easily perform comparably to the strictly non-blocking switch by having slightly more fibers per link. Also, such a problem can be tackled by several different approaches. The most efficient one is to equip the switch with more buffers rather than to increase the switching-board size or to enhance the buffer sharing between different switching boards.     

Optical Interconnection Networks for Terabit Packet Switches

The challenge of building packet switches with terabit capacity is being met by wavelength division multiplexing (WDM) where the benefits of optical fiber are exploited. Two kinds of WDM-based bufferless optical interconnection networks are proposed in this paper to interconnect multiple electronic packet switch modules. One is based on 3-stage Clos principle and the other is based on broadcast-and-select principle. The proposed optical interconnection networks are implemented with small modular structures to provide capacities in the range of terabit per second. Their architectures, component and interconnection complexity, and power budget analyzes are presented. In addition, the crosstalk caused by the finite ON-OFF ratio of semiconductor optical amplifier is discussed. Bit error rates with respect to different ON-OFF ratios and extinction ratios are also evaluated. It is concluded that it is feasible to implement optical interconnection networks by using state-of-the-art WDM technology, and they are excellent candidates for future terabit packet switching systems.     

Multiclass scheduling algorithms for the DAVID metro network

The data and voice integration over dense wavelength-division-multiplexing (DAVID) project proposes a metro network architecture based on several wavelength-division-multiplexing (WDM) rings interconnected via a bufferless optical switch called Hub. The Hub provides a programmable interconnection among rings on the basis of the outcome of a scheduling algorithm. Nodes connected to rings groom traffic from Internet protocol routers and Ethernet switches and share ring resources. In this paper, we address the problem of designing efficient centralized scheduling algorithms for supporting multiclass traffic services in the DAVID metro network. Two traffic classes are considered: a best-effort class, and a high-priority class with bandwidth guarantees. We define the multiclass scheduling problem at the Hub considering two different node architectures: a simpler one that relies on a complete separation between transmission and reception resources (i.e., WDM channels) and a more complex one in which nodes fully share transmission and reception channels using an erasure stage to drop received packets, thereby allowing wavelength reuse. We propose both optimum and heuristic solutions, and evaluate their performance by simulation, showing that heuristic solutions exhibit a behavior very close to the optimum solution.     

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.     

Buffering management schemes for optical variable length packets under limited packet sorting

This paper studies several buffering strategies for optical packet switching (OPS) under limited packet sorting. Three schemes, which are able to sort newly arrived packets based on packet’s length as well as capability of finding the minimum buffer occupancy, are analyzed and compared. Results show that all three proposed schemes could improve OPS performance considerably in terms of probability of packet loss (PPL) and probability of information loss (PIL). In addition, the simulation results show that not all the newly arrived packets need to be sorted in order to obtain minimum packet loss probability. Since the amount of packets and thus the packet processing time is significant in OPS, it is possible that not all the packets can be processed using one of the buffering strategies. An important finding of this paper is that if only 10% of the packets are sorted, the PPL is comparable to the minimum packet loss value obtained when 100% of the packets are sorted.     

Contribution to future metropolitan access network planning considering link failures in photonic switched networks

Photonic switching for traffic optimization in Metropolitan Access Optical Networks using technologies of optical packet/burst switching (OPS/OBS) seems to be a future-proof solution considering high capacity and the highly variable traffic in these networks. Looking for this future, we propose a study based on the analyze of survivability of traffic unconsidering protection mechanism but considering mesh and ring efficient topologies with node architecture to OPS. The results are obtained through computer modeling and simulation. The main parameters analyzed are as follows: capacity, average number of hops, packet loss fraction, and link utilization for each topology. These parameters are adopted to evaluate the performance of each topology considering failures. Minimum electronic buffering is included at node ingress (client side) to avoid packet loss in the access to the optical layer: high throughput and low latency. It is also observed that optical buffering is not necessary for adequate network performance.     

The study of QoS guarantee in the optical burst switching internet backbone

Optical switching technology can be categorized into optical circuit switching (OCS), optical packet switching (OPS) and optical burst switching (OBS). OCS is suitable for large amounts of data transmission; however, the channel utilization is inefficient when the traffic flows are intermittent. OPS can be easily adapted to any higher layer and is suitable for bursty traffic, but it requires a highly complex technology and optical buffer. The new switching paradigm, OBS, can provide higher bandwidth utilization and meanwhile avoid the complexity in OPS technology.In this paper, we investigate how the quality of service (QoS) can be guaranteed and reliable transmission can be supported in the OBS-based Internet backbone. We propose the adjustable-time-counter-based (ATCB) burst assembly and the non-real time packet retransmission mechanisms and apply them in the ingress router of the OBS Internet backbone to guarantee the quality of real time applications and lossless requirement of non-real time services. Moreover, traffic shaped is performed for real time packets in the egress router so that the real time property is preserved with a low jitter. Simulation results show that the burst blocking probability using the ATCB burst assembly is improved, compared with the time-counter-based (TCB) and burst-length-threshold-based (BLTB) mechanisms. The delay, loss and jitter of real time service conform to the QoS requirement. Meanwhile, the delay of non-real time service also falls in the acceptable range.     

Performance of fiber delay‐line buffers in asynchronous packet‐based optical switching networks with wavelength conversion

A major challenge in asynchronous packet‐based optical networks is packet contention, which occurs when two or more packets head to the same output at the same time. To resolve contention in the optical domain, two primary approaches are wavelength conversion and fiber delay line (FDL) buffering. In wavelength conversion, a contending packet can be converted from one wavelength to another in order to avoid conflict. In FDL buffering, contending packets can be delayed for a fixed amount of time. While the performance of wavelength conversion and FDL buffering has been evaluated extensively in synchronous networks with fixed‐sized packets, in this paper, we study the performance of FDL buffers in asynchronous packet‐based optical networks with wavelength conversion. An analytical model is proposed to evaluate the performance in terms of packet loss probability and average delay. Extensive simulation and analytical results show that, with appropriate settings, FDL buffers can perform much better in switches with wavelength conversion than in switches with no conversion. Copyright © 2014 John Wiley & Sons, Ltd.     

Restoration of IP Networks by Using a Hybrid Interacting Mechanism Between Layer 2 & 3 in the Networks Over OA&M

The Internet generation infrastructure is growing and moving towards a consistency model of high-speed nodes interconnected by optical core networks. Therefore, the interaction between IP and optical network layers, specifically, the routing and signalling aspects, are maturing and they can lead to enhance the services and network performance. At the same time, a consensus has emerged in the industry on utilizing IP-based protocols for the optical control plane. This paper defines a new technique for IP over Optical networks (interacting between bottom layers), considering both the IP-based control plane for optical networks as well as IP-optical network interactions (together referred to as “IP over optical networks”). In this paper, a new technique is proposed that reduces recovery time by making an interoperation between the data link layer (layer 2) with the network layer (layer 3). The proposed technique enables layer 2 with layer 3 to make a hybrid mechanism that improves the recovery time in the network through two cases (1st-Detection and 2nd-Rerouting). The technique involves detecting failure in less time through layer 2 and rerouting traffic through an alternative path between source and destination by using a new algorithm in layer 3 to deliver data packets without the need to wait for the routing protocol to update the network topology and compute the routing table. In case of the network recovery, layer 2 has demonstrated its capability to detect failure extremely quickly shown through the immediate detection of the loss of signals for the link or node. The recovery mechanism, i.e., the rerouting mechanism, which leads the node, switches the data packets through an adjacent node to its destination via the life node. The latter is created by the proposed mechanism before the occurrence of the failure. The aim of this mechanism is to avoid loss of packets, improve QoS and improve recovery time as we have shown in the results shown below.     

Scheduling algorithms for optical packet fabrics

Utilizing optical technologies to build packet fabrics for high-capacity switches and routers has several advantages in terms of scalability, power consumption, and cost. However, several technology related problems have to be overcome to be able to use such an approach. The reconfiguration times of optical crossbars are longer than those of electronic fabrics and end-to-end clock recovery in such systems add to the reconfiguration overheads. Both these problems can limit the efficiency of optical packet fabrics. In addition, existing work on input-buffered switches mostly assumes fixed size packets (referred as envelopes in this paper). When fixed size switching is used for Internet protocol networks where packets are of variable size, the incoming packets need to be fragmented to fit the fixed size envelopes. This fragmentation can lead to, possibly large loss of bandwidth and even instability. This paper addresses all of the above issues by presenting packetization and scheduling techniques that allow optical packet fabrics to be used within switches and routers. The proposed scheme aggregates multiple packets in a single envelope and when used in combination with proper scheduling algorithms, it can provide system stability as well as bandwidth and delay guarantees. As a result of the aggregation method, the reconfiguration frequency required from the optics is reduced, facilitating the use of optical technologies in implementing packet switch fabrics.     

The Knockout Switch: A Simple, Modular Architecture for High-Performance Packet Switching

A new, high-performance packet-switching architecture, called the Knockout Switch, is proposed. The Knockout Switch uses a fully interconnected switch fabric topology (i.e., each input has a direct path to every output) so that no switch blocking occurs where packets destined for one output interfere with (i.e., block or delay) packets going to different Outputs. It is only at each output of the switch that one encounters the unavoidable congestion caused by multiple packets simultaneously arriving on different inputs all destined for the same output. Taking advantage of the inevitability of lost packets in a packet-switching network, the Knockout Switch uses a novel concentrator design at each output to reduce the number of separate buffers needed to receive simultaneously arriving packets. Following the concentrator, a shared buffer architecture provides complete sharing of all buffer memory at each output and ensures that all packets are placed on the output line on a first-in first-out basis. The Knockout Switch architecture has low latency, and is self-routing and nonblocking. Moreover, its Simple interconnection topology allows for easy modular growth along with minimal disruption and easy repair for any fault. Possible applications include interconnects for multiprocessing systems, high-speed local and metropolitan area networks, and local or toll switches for integrated traffic loads.     

QoS Differentiation in Asynchronous Bufferless Optical Packet Switched Networks

Optical Packet/Burst Switched networks (OPS/OBS) have emerged as promising all-optical network architectures for future core networks due to good resource utilization and the ability to handle high line speeds in a cost-efficient way. Regarding these networks, Quality of Service (QoS) differentiation should be provided in order to give sufficient QoS to real-time applications, and to utilize network resources optimally. In this paper, we present three fundamental different QoS differentiation schemes suitable for asynchronous bufferless OPS. We present analytical models and derive explicit results for the packet loss rates. We then present a general evaluation framework followed by a comparison study of the considered QoS differentiation schemes. Simulation and analytical results show that QoS differentiation schemes based on preemption have better performance than drop based and access-restriction based QoS differentiation schemes.     

Performance analysis of a bufferless OBS node considering the streamline effect

In this letter, we describe and analyze a phenomenon unique to bufferless optical burst switched (OBS) networks called the streamline effect. This is the phenomenon wherein bursts within an input stream to a core node only contend with those from other input streams but not among themselves. It causes the burst loss probabilities at a link to depend strongly on the number of input streams to the link and their relative burst rates. We derive a burst loss probability formula that can account for all the observed burst loss behaviors. The accuracy of the analytical formula is confirmed through various simulation scenarios.