Optical Packet Research Articles

Multidomain Network Hypervisor for Abstraction and Control of OpenFlow-Enabled Multitenant Multitechnology Transport Networks [Invited

A virtualization architecture is presented for deploying multitenant virtual networks on top of multitechnology optical networks. A multidomain network hypervisor (MNH) and a multidomain SDN orchestrator (MSO) are introduced with this purpose. TheMNHprovides an abstract view of each virtual network and gives control of it to each independent customer SDN controller. The MNH is able to provide virtual networks across heterogeneous control domains (i.e., generalized multiprotocol label switching and OpenFlow) and transport technologies (i.e., optical packet switching and elastic optical networks). The MSO is responsible for providing the necessary end-toend connectivity. We have designed, implemented, and experimentally evaluated the MNH and MSO in an international testbed across Spain, UK, Germany, and Japan.

Scalable two- and three-dimensional optical labels generated by 128-port encoder/decoder for optical packet switching.

This paper deals with massive number of optical code (OC) label generation and recognition for scalable optical packet switching (OPS) networks. In order to expand the system scalability of code label processing, we develop a record port count 128 x 128 optical encoder/decoder (E/D) and propose a novel three-dimensional (3-D) optical label combining code label with wavelength and polarization. In the experiment, we conduct a proof-of-concept demonstration of 4-code x 2-wavelength x 2-polarization and validate that the 3-D labeling scheme can consequently increase the available number of code label up to more than 1,000 labels. Real-time labeling performance using a field programmable gate array (FPGA)-based processor and crosstalk influence at an optical switch are also experimentally evaluated.

On-Chip Reflectivity-Tunable Single Wavelength Reflector

We propose an on-chip reflectivity-tunable single wavelength reflector, based on a coupling-modulation microring with an integrated reflective element. Reflectivity ranging from 0 to 1 can be obtained by tuning the coupling coefficient. With the reflective element parameters chosen carefully, the operation bandwidth of the reflector can be designed to accommodate the input signal with different bit rates and modulation formats. Combining the proposed reflector with delay-tunable photonic elements, an on-chip reconfigurable optical correlator is proposed. This on-chip optical correlator demonstrates reconfigurability, scalability, and size reduction, and thus, it is promising for on-chip optical packet switching networks.

Optimal Virtual Slice Composition Toward Multi-Tenancy Over Hybrid OCS/OPS Data Center Networks

Multi-tenancy is a key feature of modern data centers. It allows for the existence of multiple independent virtual infrastructures, called virtual slices, on top of the same physical infrastructure, each of them specially tailored to the tenants' needs. In such a scenario, an optimal mapping of the virtual slices plays a capital role toward an efficient utilization of the data center network resources, potentially saving costs for the data center owner. However, due to the increasing trend of bringing optics to data center networks, specific virtual slice mapping mechanisms accounting for the particularities of the optical medium (e.g., wavelength continuity constraint) have to be investigated. For this, we present an integer linear programming (ILP) model for optimally mapping a set of virtual slices from different tenants in a hybrid optical circuit switching (OCS)/optical packet switching (OPS) data center network with the aim to minimize the necessary optical transponders to be equipped in the network. Additionally, we also present a lightweight heuristic for the cases in which the ILP model scalability is compromised. The benefits of the proposals are highlighted by benchmarking them against a pure OCS solution through extensive simulations.

Queueing analysis of optical and electronic combined buffer for optical packet switches

Optical packet switching provides high-speed and energy-efficient packet forwarding. Fiber delay line (FDL)-based buffer is one practical solution to avoid packet contention in optical packet switches. However, the FDL buffer suffers from scalability problems, and thus, cannot accommodate a large volume of network traffic at a time. An optical and electronic combined (OE) buffer architecture alleviates the scalability issues. This architecture uses supplementary electronic RAM buffers when the traffic volume increases. We then propose an analytical model of the OE buffer to estimate the blocking probability and the average delay. We find that complex OE combined model can be simply modeled by combination of an existing analytic model of the FDL buffer and an M/M/1/K queueing system. Comparing analytical results with packet-level simulation results, we demonstrate that our analytical results display characteristics similar to the simulation results. This analysis enables to determine appropriate setting of the OE buffer such as the buffer size and use policy.

Mode-selective optical packet switching in mode-division multiplexing networks.

A novel mode-selective optical packet switching, based on mode-multiplexers/demultiplexers and multi-port optical micro-electro-mechanical systems (MEMS) switches, has been proposed and experimentally demonstrated. The experimental demonstration was performed using the LP(01), LP(11a) and LP(11b) modes of a 30-km long mode-division multiplexed few-mode fiber link, utilizing 40 Gb/s, 16-QAM signals.

Delay-differentiated scheduling in optical packet switches for cloud data centers

We consider differentiated timecritical task scheduling in a N×N input queued optical packet switch to ensure 100% throughput and meet different delay requirements among various modules of data center. Existing schemes either consider slot-by-slot scheduling with queue depth serving as the delay metric or assume that each input-output connection has the same delay bound in the batch scheduling mode. The former scheme neglects the effect of reconfiguration overhead, which may result in crippled system performance, while the latter cannot satisfy users’ differentiated Quality of Service (QoS) requirements. To make up these deficiencies, we propose a new batch scheduling scheme to meet the various portto- port delay requirements in a best-effort manner. Moreover, a speedup is considered to compensate for both the reconfiguration overhead and the unavoidable slots wastage in the switch fabric. With traffic matrix and delay constraint matrix given, this paper proposes two heuristic algorithms Stringent Delay First (SDF) and m-order SDF(m-SDF) to realize the 100% packet switching, while maximizing the delay constraints satisfaction ratio. The performance of our scheme is verified by extensive numerical simulations.

Optical and Electronic Combined Buffer Architecture for Optical Packet Switches

Optical packet switch systems with fiber delay line (FDL) buffers provide high-throughput, bit-rate independent, energy-efficient, and transparent forwarding. However, the FDL buffer experiences scalability issues and, thus, cannot accommodate a large amount of network traffic. In this paper, we first investigate network performance of the FDL buffer. Due to the discrete-time nature of the FDL buffer, throughput of Transmission Control Protocol (TCP) applications is not improved drastically in spite of the addition of extra FDLs. Then, we propose optical and electronic combined buffer architecture. The combined buffer is composed of an FDL buffer and a supplementary electronic buffer having power management functions according to its occupancy. The combined buffer uses only the FDL buffer and keeps its energy efficiency when traffic volume is low. In doing so, this architecture reduces power consumption to 70% of that of a buffer composed only of electronic devices under non-peak traffic. As the traffic volume increases, the combined buffer turns on its electronic devices and accommodates the increased traffic. Simulation results reveal that the proposed buffer architecture obtains 1.5 times larger throughput than that of the existing FDL buffer against suddenly increased traffic. Finally, we show that the supplementary electronic components have to be ready to use in less than 300 ms.

Exploring parallelism and desynchronization of TCP over high speed networks with tiny buffers

The buffer sizing problem is a big challenge for high speed network routers to reduce buffer cost without throughput loss. The past few years have witnessed debate on how to improve link utilization of high speed networks where the router buffer size is idealized into dozens of packets. Theoretically, the buffer size can be shrunk by more than 100 times. Under this scenario, widely argued proposals for TCP traffic to achieve acceptable link capacities mandate three necessary conditions: over-provisioned core link bandwidth, non-bursty flows, and tens of thousands of asynchronous flows. However, in high speed networks where these conditions are insufficient, TCP traffic suffers severely from routers with tiny buffers.To explore better performance, we propose a new congestion control algorithm called Desynchronized Multi-Channel TCP (DMCTCP) that creates a flow with parallel channels. These channels can desynchronize each other to avoid TCP loss synchronization, and they can avoid traffic penalties from burst losses. Over a 10 Gb/s large delay network ruled by tiny buffer routers, our emulation results show that bottleneck link utilization can reach over 80% with much fewer number of flows. Compared with other TCP congestion control variants, DMCTCP can also achieve much better performance in high loss rate networks. Facing the buffer sizing challenge, our study is a new step towards the deployment of optical packet switching networks.

Using Stuffed Quadratic Congruence Codes for SAC Labels in Optical Packet Switching Network

While optical codes in optical code-division multiple access (OCDMA) has been extensively investigated, research of optical coding labels in optical multi-protocol label switching (MPLS) network is relatively unexplored. This paper studies optical labels with spectral-amplitude coding (SAC) for optical packet switching (OPS) network. The research focuses on implementing two different codes in optical labels: stuffed quadratic congruence (SQC) codes and M-sequence codes. Label-error rate (LER) is quantified to ascertain the influence of code pattern on system performance. The findings suggest that SQC coded label can support more numbers for label stacking, due to its very low cross-correlation value.

SDN-Enabled OPS With QoS Guarantee for Reconfigurable Virtual Data Center Networks

Optical packet switching (OPS) can enhance the performance of data center networks (DCNs) by providing fast and large-capacity switching capability. Benefiting from the software-defined networking (SDN) control plane, which could update the look-up-table (LUT) of the OPS, virtual DCNs can be flexibly created and reconfigured. In this work, we have implemented and assessed an SDN-based control framework for an OPS node, where the OpenFlow protocol has been extended in support of the OPS switching paradigm. Application flows are switched by the OPS at submicrosecond hardware speed, decoupled from the slower (millisecond timescale) SDN control operation. By the DCN infrastructure provider, the virtual networks become directly programmable with the abstraction of the underlying OPS node. Experimental results validate the successful setup of virtual network slices for intra-data center interconnect and quality of service (QoS) guarantee for high-priority application flows. Data plane resources are efficiently shared by exploiting statistical multiplexing. In addition, the capability of exposing per-port OPS traffic statistics information to the SDN controller enables the implementation and experimental validation of load balancing algorithms to improve the QoS performance.

Application-based dynamic reconfiguration in optical network-on-chip

We propose a new optical reconfigurable Network-on-Chip (NoC), named ReFaT ONoC (Reconfigurable Flat and Tree Optical NoC). ReFaT is a dynamically reconfigurable architecture, which customizes the topology and routing paths based on the application characteristics. ReFaT, as an all-optical NoC, routes optical packets based on their wavelengths. For this purpose, we propose a novel architecture for the optical switch, which eliminates the need for optical resource reservation, and thus avoids the corresponding latency and area overheads. As a key idea for dynamic reconfiguration, each application is mapped to a specific set of wavelengths and utilizes its dedicated routing algorithm. We compare the proposed reconfigurable optical NoC with traditional electrical NoCs, as well as non-reconfigurable optical NoCs in terms of data transmission delay, power consumption, and energy dissipation. The simulation results show that on average, ReFaT reduces power and energy consumption by 54.75% and 40%, respectively, compared to the non-reconfigurable ONoC.

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

Optical packets contention in optical packet switched (OPS) networks can lead to Optical packet loss which will reduce network performance. In this paper, a new contention avoidance technique is proposed which utilizes the combination of traffic shaping at ingress switches and a time slot reservation technique using the coordination of egress switches. This novel protocol is called egress coordination OPS (EGCOPS) suitable for buffer-less slotted-OPS networks. Simulation results show that the EGCOPS outperforms the original slotted-OPS protocol in terms of optical packet loss rate (PLR), with or without using wavelength conversion. However, the packet shaping feature of EGCOPS can increase the delay of users’ packets at the queues of ingress switches. There is a trade-off between the PLR improvement and the additional delay applied to the users’ packets. Still, EGCOPS parameters can be selected in such a way that additional delay can be tolerable by users’ applications.

A novel contention resolution scheme based on network coding for optical multicast node

Optical buffering using fiber delay lines (FDLs) has been proved to be the best choice to resolve contention for optical packet switching node. Due to the absence of optical RAM, XOR network coding becomes a new packet contention resolution technology in the all-optical networks. In this paper, we present an all-optical node architecture configured with a primary network coding module and a supplementary FDL buffer module for contention resolution. The contending packet is resolved by output shared XOR network coding module firstly without any time delay. The feedback shared FDL buffer is used to resolve the still packet contention after NC module, which is constructed by a FDL buffer loop group. Moreover, the minimum length-cascaded buffer scheduling technique (MLCBST) is designed to schedule the contending packet to available output with minimal transmission delay. The simulation results show that the proposed architecture and the MLCBST have an obvious effect on reducing the packet loss probability and decreasing the average delay for optical packet switching node.

Physical layer analysis of AWG based optical packet switch architecture

Optical packet switching is the next generation switching technique in conjunction with mature electronic switching. In optical packet switching, the core router plays an important role in the routing and buffering of packets. In recent past, AWG based optical router are proposed and heavily investigated due to its very low insertion loss and very simple wavelength dependent routing pattern. In this paper, a further modification in an AWG based optical packet switch is proposed. These modifications improve the signal reception quality at the output. Power budget analysis is presented to evaluate the buffer capacity. As a physical layer parameters, the minimum switch power is evaluated using the mathematical model and at the BER level of 10−9 the maximum buffering capacity is evaluated. Finally, simulation results are presented in terms of network layer parameters, packet loss probability and average delay. The packet loss probability is as low as 10−7 well under acceptable limits.

All-optical 2-bit header recognition and packet switching using polarization bistable VCSELs.

We propose and evaluate an all-optical 2-bit header recognition and packet switching method using two 1.55-µm polarization bistable vertical-cavity surface-emitting lasers (VCSELs) and three optical switches. Polarization bistable VCSELs acted as flip-flop devices by using AND-gate operations of the header and set pulses, together with the reset pulses. Optical packets including 40-Gb/s non-return-to-zero pseudo-random bit-sequence payloads were successfully sent to one of four ports according to the state of two bits in the headers with a 4-bit 500-Mb/s return-to-zero format. The input pulse powers were 17.2 to 31.8 dB lower than the VCSEL output power. We also examined an extension of this method to multi-bit header recognition and packet switching.

A novel delay line buffering architecture for asynchronous optical packet switched networks

Optical buffering is one major challenge in realizing all-optical packet switching. In this paper we focus on a delay-line buffer architecture, named a Multiple-Input Single-Output (MISO) optical buffer, which is realized by cascaded fiber delay lines (FDLs). This architecture reduces the physical size of a buffer by up to an order of magnitude or more by allowing reuse of its delay line elements. We consider the MISO buffers in a network scenario where the incoming packets are asynchronous and of fixed length. A novel Markov model is developed to analyze the performance of our buffering scheme, in terms of packet loss ratio, average packet delay and the output link utilization. Both simulation and analytical results show that the length value of basic FDL element will significantly affect the performance of this buffer. This paper gives clear guidelines for designing optimal basic FDL lengths under different network scenarios. It is noticeable that this optimal length value is independent of the buffer sizes for specific traffic load and pattern.

All-Optical Packet Alignment Using Polarization Attraction Effect

We demonstrate submicrosecond polarization alignment of nonreturn to zero 10-Gbit/s optical packets using an all-optical polarization attraction effect in a low polarization-mode dispersion-shifted fiber. We experimentally demonstrate the alignment of optical packets having durations of 205 and 409 ns using a counter-propagating four-wave mixing process. To evaluate this technique, consecutive packets were aligned with distinctly different polarization states using a fiber interferometer. We experimentally demonstrate the recovery of optical packets that were lost due to their state of polarization (SOP) misalignment with the receiver. Polarization stability of the proposed system was assessed over a period of 30 min and it was observed that the scheme was able to correct the SOP and equalize the received optical power.

Dispersion of double-slot microring resonators in optical buffer

In the optical packet switching network, optical buffer is an important device. Microring resonator optical buffers provide good delay performance and flexibility in design. By cascading multiple microring resonators, higher delay-bandwidth product is obtained, but the requirements of high integration and low dispersion are hard to satisfy simultaneously. Double-slot waveguide was proposed to construct highly integrated racetrack microring resonators in this study. Based on dispersion analysis of the thickness of each layer of a waveguide, the structure of waveguide was optimized to reach flat and low dispersion. Average dispersions of straight and 3 μm bend waveguides were 5.1 ps/(nm∙km) and 4.4 ps/(nm∙km), respectively. Besides, the additional loss from coupling was greatly reduced when applying proper relative displacement between straight and bend waveguides. Theoretical and design basis provided in this paper will help to develop multi-microring optical buffers in the future.

Torus-Topology Data Center Network Based on Optical Packet/Agile Circuit Switching with Intelligent Flow Management

We review our work on an intra-data center (DC) network based on co-deployment of optical packet switching (OPS) and optical circuit switching (OCS), conducted within the framework of a five-year-long national R&D program in Japan (∼March 2016). For the starter, preceding works relevant to optical switching technologies in intra-DC networks are briefly reviewed. Next, we present the architecture of our torus-topology OPS and agile OCS intra-DC network, together with a new flow management concept, where instantaneous optical path on-demand, so-called <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Express Path</i> is established. Then, our hybrid optoelectronic packet router (HOPR), which handles 100 Gbps (25 Gbps × 4-wavelength) optical packets and its enabling device and sub-system technologies are presented. The HOPR aims at a high energy-efficiency of 0.09 [W/Gbps] and low-latency of 100 ns regime. Next, we provide the contention resolution strategies in the OPS and agile OCS network and present the performance analysis with the simulation results. It is followed by the discussions on the power consumption of intra-DC networks. We compare the power consumption and the throughput of a conventional fat-tree topology with the <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">N</i> -dimensional torus topology. Finally, for further power saving, we propose a new scheme, which shuts off HOPR buffers according to the server operation status.