Contention avoidance in bufferless slotted optical packet switched networks with egress switch coordination

The wavelength conversion is regarded as an effective way to resolve the optical packet contention in the wavelength domain for optical packet switching. An optical packet switching node, based on shared-per-node equipped with limited range wavelength converters and parametric wavelength converters (SPN-LP), is designed to further reduce optical packet loss probability. A novel optical packet contention resolution with priority differentiation wavelength scheduling algorithm to support quality of service (QoS) for the SPN-LP architecture is put forward in the article. The simulation results show that proposed optical packet resolution enables a good QoS differentiation, namely the high priority contending optical packet has the sufficient low packet loss probability.

We develop an integrated optical packet and circuit switching node in which signaling via optical packet switching is implemented. We demonstrate 80 (8λ × 10) Gbit/s optical packet switching with <;1E-4 packet-loss-rate, 8-lightpaths establishment, and dynamic wavelength-resources control.

An optical router with multistage distributed management features for the asynchronous optical packet switching (OPS) network is presented, which can improve switching capacity and all-optical scalability. A compact recycling-fiber-delay-line (Rec-FDL) based collision resolution mechanism is proposed to resolve the contentions for asynchronous and variable length optical packets. The analysis models of stabilities, packet loss rates (PLR) and average packet waiting latencies (PWL) for the router are developed based on the timer based optical packet assembly algorithm. The simulation shows that PLR and PWL for a 400-byte optical packet transmitted in the 32 wavelengths dense wavelength division multiplexing (DWDM) system equal to 3.48 × 10?4 and 0.072 ns, respectively. The non-blocking switching can be realized for the packets with lengths less than the buffer granularity of the Rec-FDL, and the optimized performance for the proposed router can be obtained through properly selecting of the system parameters.

This paper proposes a scheme for pump-wavelength selection in an optical packet switch (OPS) with parametric wavelength converters (PWCs). In this scheme, pump wavelengths can be dynamically changed in all time slots, and more than one PWC is allowed to convert a wavelength in a chained manner. The objective of this scheme is to minimize the packet loss rates of the OPS. This scheme is called dynamic pump-wavelength selection with chained parametric wavelength conversion (DPS-C). DPS-C selects the pump wavelength for each PWC so as to maximize the number of successful requests, which refer to packets that wish to be forwarded to their requested output fibers. A PWC, which has the advantage of multiple wavelength conversion, uses a pump wavelength that can be flexibly chosen to define which wavelengths can be converted from/to. The set of original and converted wavelengths, where the pump wavelength is set in the middle of the two wavelengths, is called the conversion pair. Each PWC supports several conversion pairs. The OPS allows each wavelength to be converted using combinations of available conversion pairs, which are not used by any requests, from more than one PWC. A simulation shows that DPS-C outperforms the conventional scheme in terms of packet loss rates in both uniform and nonuniform traffic patterns.

The present work analyzes the performance of photonic switching networks, optical packet switching (OPS) and optical burst switching (OBS), in mesh topology of different sizes and configurations. The “lossless” photonic switching node is based on a semiconductor optical amplifier, demonstrated and validated with experimental results on optical power gain, noise figure, and spectral range. The network performance was evaluated through computer simulations based on parameters such as average number of hops, optical packet loss fraction, and optical transport delay (Am). The combination of these elements leads to a consistent account of performance, in terms of network traffic and packet delivery for OPS and OBS metropolitan networks. Results show that a combination of highly connected mesh topologies having an ingress e-buffer present high efficiency and throughput, with very low packet loss and low latency, ensuring fast data delivery to the final receiver.

Optical Burst Switching and Packet Switching require the adoption of suitable contention resolution mechanisms at the optical nodes due to multiple packets/bursts attempting to access the shared output channel at the same time. Contentions can be solved by exploiting the space, wavelength and time domain. The many alternatives possible call for suitable scheduling policy aiming at an optimal system performance. This paper provides an assessment of the role of key parameters for an optical packet/burst contention resolution scheme and the resulting node cost estimation in terms of CAPEX, footprint and power consumption. Feasibility tradeoffs due to limited optical buffering and wavelength conversion capability are also taken into account.

Optical Packet Switching (OPS) has emerged as a promising candidate for the future all-optical core network. A crucial issue in such networks is how the packet length distribution (PLD) influences the performance under different traffic scenarios. In this paper we investigate how a deterministic, empirical and exponential PLD influences the Packet Loss Rate (PLR) in an optical packet switch in the case of a Poisson arrival process and a bursty Hyperexponential arrival process. We consider optical packet switches with and without Fibre Delay Line (FDL) contention buffers. Our results show that in the case of OPS without FDL buffering and with a Hyperexponential arrival process, the empirical PLD yields the lowest PLR. However, in the case of OPS with FDL buffering, the deterministic PLD yields the lowest PLR.

An optical packet and circuit integrated network (OPCInet) will allow diverse services, enhanced functional flexibility, and efficient energy consumption, as optical packet switching (OPS) and optical circuit switching (OCS) links are provided on the same infrastructure. In this paper, we design and develop a control system consisting of header processing for optical packets, signaling and routing for optical path setup, and wavelength resource control and link switching between OPS and OCS for the efficient use of resources. Our onboard optical packet header processor is capable of 16-bit longest prefix matching by embedding a 0.6W (5% of TCAM technology on the same condition), 200 million search-per-second (equivalent to 100 Gbps) forwarding engine LSI, and a statistical memory LSI for retaining the statistical information of processed headers. Reactive control automatically selects a packet or circuit link according to the statistical traffic information from the header processor and works with signaling for optical paths. We present an experimental demonstration of a new control system using a previously developed OPCInet testbed.

Modeling and performance analysis of a data center network (DCN) combining optical packet switching (OPS) and agile optical circuit switching (OCS) in a unified platform are presented. Hybrid optoelectronic packet routers which combine optical switching fabric and electronic buffers are used, supporting 100 Gbit/s (25 Gbit/s × 4 wavelengths) optical packets. OCS transmission is implemented by a novel “express path,” (ExP) an instant wavelength path on demand, which enables certain flows following the ExP with no packet contention, whereas ordinary OPS packets are switched with a deflection routing algorithm and buffering for contention resolution. This architecture can support OCS for large and reliability-sensitive data transferring and OPS for latency-sensitive applications. The performance of the network is investigated via numerical simulations of top-of-rack switch (ToR)-to-ToR traffic in an N-dimensional (up to 4096 node) torus topology. We have made the following observations: 1) torus topology results in high throughput and can be scaled out by increasing the dimension; 2) contention resolution between OPS and agile OCS is needed and our proposed strategies enhance the overall performance; and 3) in a lightly to moderately loaded case of OPS traffic, more than 100 ExPs can be supported simultaneously with acceptable affect on OPS.

Optical packet switch with multicast capability enables a broad range of applications in next generation optical networks. In this paper, we propose a cost-effective scheme to enhance the traffic performance of our earlier reported optical wavelength-routed multicast packet switch by modifying rather than adding multicast modules in the switch. A packet scheduling technique is introduced to substantially reduce the output contention for the switch. The effectiveness of our proposed scheme in conjunction with the scheduling technique is analyzed and validated by evaluation of packet loss performance of the modified multicast packet switch through simulations. The simulation results show that the multicast packet loss probability of the modified multicast switch can be significantly reduced without increasing the number of multicast modules by our proposed scheme. This means that to achieve an acceptable level of the multicast packet loss probability, the modified multicast switch requires fewer multicast modules in comparison with our earlier reported switch.

We propose and implement an enhanced moving average (MA) based dynamic resource allocation method for optical packet and circuit integrated (OPCI) networks. In the OPCI networks, optical packet switching (OPS) and optical circuit switching (OCS) share the total bandwidth in a fiber by allocating separate wavelength resources to them. In this work, we leverage an exponential moving average (EMA) as an enhanced MA. Every time a signaling process for path establishment or release occurs at a node, the proposed method employs the EMA of number of in-use paths to reallocate wavelengths to OCS and OPS while the conventional method employs the absolute number of in-use paths. When the EMA reaches or becomes less than a threshold value, the dynamic resource allocation is executed. As results of simulations and a verification experiment, we show that the proposed method can reduce the amount of routing processes without suffering from blocking of path establishments in comparison to the conventional method. In addition, it is expected that the proposed method can effectively allocate vaster amount of bandwidth to OPS without suffering from the blocking and achieve better system performances for OPS.

In this paper, a novel optical packet switch is proposed, which uses a set of shared fiber delay lines (FDLs) and a set of shared tunable wavelength converters (TWCs) to resolve optical packet contentions. In addition, two control algorithms, i.e., Fiber-First (FF) and Wavelength-First (WF), are proposed to schedule optical packets. Performance of the novel switch is evaluated by means of simulation experiments. Simulation results show that based on either of these two control algorithms, the switch can achieve super performance without employing a lot of FDLs and TWCs. Moreover, the performance of FF and WF is compared. Detail analyses are also given in this paper.

The huge traffic growth which is currently experienced (traffic nearly doubles every year) and foreseen to continue for several years calls for high capacity networks capable of efficiently handling various types of traffic (voice, video, data). WDM (Wavelength Division Multiplexing) optical networks exploiting also fast optical packet switching are expected to provide the required capacity and flexibility for the next generation of high-speed networks. In this context, the European ACTS KEOPS project (KEys to Optical Packet Switching) has concentrated on the definition, development and assessment of optical packet switching and routing networks capable of providing transparency to the payload bit rate. Optical packets of fixed duration have been considered where the payload bit rate, which is user-dependent, can vary from a few hundred Mb/s to 10 Gb/s whereas the header is at a fixed low bit rate (622 Mb/s) to facilitate processing at the network/node interfaces. The challenge addressed by the KEOPS project was therefore to combine the advantages of the relatively coarse-grained WDM techniques with emerging all-optical switching capabilities to yield a high-throughput optical platform directly underpinning next generation data networks (ATM, IP). The project which involved nine partners (Alcatel Alsthom Recherche (F), France Telecom-CNET (F), CSELT (I), Technical University of Denmark (DK), Alcatel CIT (F), Alcatel SEL (D), University of Bologna (I), ETH-Zurich (CH) and University of Strathclyde (UK)) has addressed both network and system studies as well as demonstration experiments based on advanced optoelectronic components developed in the project in order to validate the proposed concept.

Multiwavelength optical packet and circuit integrated (OPCI) networks featured by the agile wavelength resource adjustment have been developed. With quick adjustment of wavelength resource, e.g., in emergency cases like disasters, the capacity of optical packet switching (OPS) portion of OPCI network where is closed to the disaster area can be extended quickly first so as to accommodate the significantly increased packet traffic. Here, we are motivated to investigate the efficient way for taking advantage of the emergently extended wavelength resource (for OPS) in the most congested part of the OPCI core network (i.e., outside of but close to the disaster area). In this paper we propose a waveband (WB) assignment scheme for multiwavelength-based OPS network which mixes the dedicated WBs and shared WBs assignment to isolate the heavy traffic flows as many as possible, and minimize the packet congestion efficiently. Simulation results reveal that our proposal can significantly reduce the packet loss, hence enhance the robustness of OPCI networks in emergency situations.

We have proposed an optical packet and circuit integrated (OPCI) network which provides over Tbps capacity with low power consumption by optical technologies and deals with diversity of services from best-effort service to quality of service guaranteed service by packet and circuit switching schemes. Recently, we developed an integrated optical packet and circuit switch (OPS/OCS) node and constructed an OPCI ring network testbed. The integrated OPS/OCS node has a function to encapsulate Internet Protocol (IP) packets coming from client networks into optical packets. In this paper, we discuss about how to incorporate OPCI networks into current IP-based wide area networks. For the first time, we experimentally demonstrate an OPCI network testbed incorporated into a multihomed network where IP addresses are hierarchically allocated. The multihomed network provides multiple links with different characteristics such as a 100 Gbps optical packet link for high capacity and shared bandwidth services, a 10 Gbps optical path link for high quality and dedicated bandwidth services, and a 10Gbit Ethernet link for low cost services. Therefore, users can choose adequate link technologies according to application or traffic conditions. We successfully verify that high definition video data is efficiently transported by switching link technologies in the multihomed network.