Search

The recent rapid advancement of cloud-native networking infrastructure has leveraged the resource virtualization technology of containers to realize diverse microservice-based applications in 5G/6G networks and clouds. Containers drastically enhance the efficiency of computational resource allocation and utilization as compared to the related virtualization technology of Virtual Machines (VMs). The networking environment leveraging both VM and container mixed virtualization technologies makes the most use of them to realize a computational platform whose resources can be dynamically adjusted to a fine granularity. To continuously meet the required levels of quality of services in 5G/6G networks and clouds in that platform, an agile and autonomous data analytics system in the control plane is essential for the accurate prediction of server workloads and dynamic allocation of enough amount of computational resource. In this paper, we introduce a framework, which complies with Recommendation ITU-T Y.3177, for autonomous computational resource control and management. The framework consists of an advanced data analytics system and a resource control system. We propose an architecture for the advanced data analytics system consisting of learning and prediction components. The learning component includes a three-stage intelligent model pipelining with three cascaded machine learning models, nonlinear regression, clustering, and multiple regression. These models determine the fluctuation trends in CPU utilization, classify services with similarities in the trends, and predict the peak CPU utilization of each containerized microservice. We evaluate the proposed models through experiments and numerical analysis. The results prove that the models support agile data analytics, which can complete data processing in the time granularity of seconds and achieve higher prediction accuracy of CPU utilization.

5G communication networks will be complex due to the emergence of an unprecedented huge number of new types of connected devices and services. Moreover, the on-demand creation of virtual network slices, each suitable for a different application, is posing challenges to the efficient management of network resources, while optimally satisfying the quality of service requirements in time-varying workloads and network conditions. This article, which is tutorial in nature, introduces 5G network slices (from the point of view of the non-wireless part of the network) and elaborates the necessity of automation of network functions related to the design, construction, deployment, operation, control, and management of network slices. It revisits machine learning techniques applicable to the automation of network functions. It then presents a machine-learning-based framework for the operation and control of network slices by continuously monitoring workload, performance, and resource utilization, and dynamically adjusting the resources allocated to network slices. Preliminary results of workload prediction accuracy obtained from the analysis of real-life data collected from a web server are also reported.

The next generation communication networks are going to be composed of virtualized node and link resources, which can be divided into various slices. Different virtual network slices can be configured with different amounts of resources so that each slice can be able to optimally satisfy the quality of service requirements of the given category of Internet of Things services. Virtualized resource allocated to slices can be dynamically adjusted to adapt the network capability to varying workload. In this article, we present our ongoing research on autonomous mechanisms for the intelligent allocation and dynamic adjustment of resources to make the virtualized network system scalable while optimally utilizing the resources. We present mechanisms for achieving three types of scalability: vertical scaling by adjusting virtualized resources allocated to a network function within a node, horizontal scaling by adjusting the number of virtualized network nodes and servers, and internetwork scaling by the arbitration of limited resources among many network slices. The preliminary performance evaluation results are also discussed.

Interconnection between our existing optical packet and circuit integrated network (OPCInet) and an ecosystem for network service provisioning named ChoiceNet enables network service providers to provide diverse services and realize control driven by the providers. In this paper, we design and implement control interfaces for the interconnection, which enables each ChoiceNet provider to explicitly control both OPCInet- and OpenFlow-based networks by use of a single common interface with proprietary/easily modifiable message formats for all control functions. We construct an experimental setup between the USA and Japan, and demonstrate that the OPCInet and an OpenFlow-based network placed in Japan are successfully controlled by request from the ChoiceNet provider installed in the USA. We also show that the time required to complete Telnet access to the OPCInet C-plane for each control is no more than 150 ms over the long distance from the USA to Japan.

Service function chaining (SFC) enables network operators to flexibly provide diverse services such as Internet-of-Things (IoT) and mobile applications. SFC technologies are required to offer stable and guaranteed quality-of-service (QoS) even in the circumstances of dynamically-changing resource demands and traffic volumes. To meet QoS requirements against time-varying network environment, infrastructure providers need to dynamically adjust the amount of computational resources such as CPU assigned to virtual network functions (VNFs) in each service function chain. If the increased resource demand of a VNF cannot be satisfied from the available resources in the current node, the VNF should be migrated to another node. Related work has been tackling SFC embedding and scheduling issues by applying various techniques such as game theory, integer linear programming and heuristic methods. However, they have limitations that they cannot provide an agile operation of VNF migration as they require a large number of iterations. In order to overcome the limitations, we propose to utilize an encoder-decoder recurrent neural network and train it to solve the VNF migration scheduling problem. Through the simulation, we verify that the proposed method can agilely determine VNF redeployment locations by keeping the frequency of server overload and VNF migration minimum.

Along with rapid prevalence of Internet of things (IoT) and mobile network services, network function virtualization (NFV) based infrastructure requires dynamic computational resource adjustments in response to time-varying environments (network traffic, resource usage, failure status, etc). To provide agile resource control and adaptiveness, it is effective to predict a virtual server load by means of machine learning technologies for proactive control. In this paper, we propose a regression analysis model utilizing a sparse modeling to predict the average server load in a future specific time period on the basis of the server load in a past specific time period. We target at the number of access to a Web server as a virtual server load. The model can reduce the prediction error compared to a traditional least square method. Besides, we propose two methods to further reduce prediction errors: (A) an algorithm for determining the time period targeted for explanatory variables, and (B) two-stage regression analyses. As a result of MATLAB calculations, we show that application of the sparse modeling can reduce the prediction error by more than 40%, and besides, the above two proposed methods are effective at further reducing the prediction error with a few minutes’ learning time. Our model also contributes to make humans’ posterior analysis easier by making a short list of explanatory variables of data than conventional methods.

Software defined networking (SDN) and network function virtualization (NFV) technologies enable us to create, manage and operate network services more flexibly. Distributed placement of controllers reduces the response time of control operations and distributes load of controllers. Many prior studies have investigated several algorithms of controller placement in networks. However, those algorithms take longer computation time and require a lot of information. Thus, they are difficult to use, especially in large networks. To simplify the controller placement problem, we propose a salience-based approach, where the salience value of a link is defined as the number of shortest path trees that traverse the link. We found that by placing controller functions on the nodes having many high-salient links, the controller placement approach shortens the length of control paths and keeps controllers' load low even in large networks.

Machine learning has the capability to provide simpler solutions to complex problems by analyzing a huge volume of data in a short time, learning for adapting its functionality to dynamically changing environments, and predicting near future events with reasonably good accuracy. The 5G communication networks are getting complex due to emergence of unprecedentedly huge number of new connected devices and new types of services. Moreover, the requirements of creating virtual network slices suitable to provide optimal services for diverse users and applications are posing challenges to the efficient management of network resources, processing information about a huge volume of traffic, staying robust against all potential security threats, and adaptively adjustment of network functionality for time-varying workload. In this paper, we introduce about the envisioned 5G network slicing and elaborate the necessity of automation of network functions for the design, construction, deployment, operation, control and management of network slices. We then revisit the machine learning techniques that can be applied for the automation of network functions. We also discuss the status of artificial intelligence and machine learning related activities being progressed in standards development organizations and industrial forums.

We have previously presented ChoiceNet, a choice-based architecture for the future Internet. ChoiceNet enables users to choose and pay for advertised competing service component offerings and independently compose the end-to-end service they require. Over time, this system rewards successful services, enabling innovation at all levels of networking, and creating a beneficial ecosystem. Provisioning is an internal operation performed by the service providers. In this paper, we examine practical ways of handling the provisioning of ChoiceNet services and discuss the benefits and disadvantages of each approach.

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.

This paper presents and experimentally demonstrates the generalized architecture of Openflow-controlled optical packet switching (OPS) network. Openflow control is enabled by introducing the Openflow/OPS agent into the OPS network, which realizes the Openflow protocol translation and message exchange between the Openflow control plane and the underlying OPS nodes. With software-defined networking (SDN) and Openflow technique, the complex control functions of the conventional OPS network can be offloaded into a centralized and flexible control plane, while promoted control and operations can be provided due to centralized coordination of network resources. Furthermore, a contention-aware routing/rerouting strategy as well as a fast network adjustment mechanism is proposed and demonstrated for the first time as advanced Openflow control to route traffic and handle the network dynamics. With centralized SDN/Openflow control, the OPS network has the potential to have better resource utilization and enhanced network resilience at lower cost and less node complexity. Our work will accelerate the development of both OPS and SDN evolution.

We have previously presented ChoiceNet, a choicebased architecture for the future Internet. ChoiceNet enables users to choose and pay for advertised competing service component offerings and independently compose the end-to-end services they require. In this paper, we investigate whether we can achieve collaborative optimization (co-optimization) of the network for both the customers and service providers through pre-grooming pricing algorithms within a flexible optical choice-based network. We evaluate the network performance of different path provisioning strategies along with a range of responsive pre-grooming pricing algorithms, and assess whether the network provider and their customers reached co-optimization, by analyzing their impact on the network’s blocking rate of circuit calls, link utilization, and its generated revenue.

In this paper, we investigate the effect of which different pricing algorithms have on a flexible optical choice-based network. Our pricing algorithms attempts to perform proactive collaborative optimization by offering some network service choice, as more attractive than others, and thus encourage users to buy services that are in surplus instead of those which use resources that are viewed more precious due to their scarcity. We evaluate these pricing algorithm's performance on the system by analyzing the impact on the network utilization, customer utilization, and the generated revenue.

We demonstrate a failure-tolerant seed lightwave distribution system by using commercially-available ROADMs and a customized centralized controller. Ability of coherent communication with low DSP-complexity can be quickly and automatically restored against seed lightwave loss.

Network virtualization techniques enable network operators to implement and provide multiple virtual networks (VNs) on a common substrate network infrastructure. The network resources should be able to be dynamically reallocated among VNs or migrated from a place to another one in various situations, such as to construct a new urgent VN in case of occurrence of some urgent contingency, or to satisfy quality of service (QoS) requirements of non-urgent VNs in case of time-varying network traffic conditions as much as possible. In this paper, we propose methods to automatically and dynamically select and migrate resources of non-urgent VNs. In particular, our proposal applies reinforcement learning for the selection of resources from alternate places to satisfy the QoS requirements. We evaluate the performances in terms of the satisfaction level of QoS requirement with various frequency of network traffic variation in a given duration and learning parameter configurations. Simulation results show that the proposed dynamic resource migration method can increase the number of times that non-urgent VNs' QoS requirements are satisfied in comparison with a static resource assignment method. Moreover, we show that, depending on traffic variation and parameter configuration, applying reinforcement learning can increase the number of times the QoS requirement is satisfied compared to the dynamic method with completely random resource selection.

Software Defined Networking (SDN) technology enables us to configure switches in a network flexibly. Previously, distributed SDN control methods have been discussed to improve their scalability and robustness. Distributed placement of controllers reduces computation load of each controller and backing up each other enhances robustness. However, these techniques do not include an emergency measure against large-scale failures such as network separation induced by disasters. In this paper, we design emergency reconfiguration of distributed SDN especially in control plane (C-plane). In our reactive approach, each node detects disconnection to the main controller. Then, a controller deployed on one of the disconnected nodes becomes the substitute controller to which the others reconnect. Meanwhile, C-plane should be reconstructed when network connectivity recovers in case many controllers exist. The main and substitute controllers detect networks are reconnected, and merge their C-planes to avoid conflict. Simulation results reveal that our proposed method recovers C-plane logical connectivity with probability of about 90% when regional failure occurs in 100 node networks. Furthermore, we show that convergence time of our reconstruction mechanism is proportional to network size.

An optical packet and circuit integrated network (OPCInet) provides both high-speed, inexpensive services and deterministic-delay, low-data-loss services according to the users’ usage scenarios, from the viewpoint of end users. From the viewpoint of network service providers, this network provides large switching capacity with low energy consumption, high flexibility, and efficient resource utilization with a simple control mechanism. This paper presents the recent progress made in the development of OPCInet and its extension to software-defined networking (SDN). We have developed OPCI nodes, which are capable of layer 3 switching from/to an Ethernet frame to/from an optical packet in the optical packet edge part and a burst-tolerant optical amplifier and an optical buffer with optical fiber delays in 100 Gbps optical packet switching part. The OPCI node achieves a packet error rate less than $10^{-4}$ and is used as a node in a lab-network that has access to the Internet. A distributed automatic control works in a control plane for the circuit switching part and in a moving boundary control between optical packet resources and circuit resources. Our optical system for packet and circuit switching works with a centralized control mechanism as well as a distributed control mechanism. We have shown a packet-based SDN system that configures mapping between IP addresses and OPCI node identifiers and switching tables according to the requests from multiple service providers via a web interface.

Network virtualization techniques enable network service providers to implement and provide multiple logical networks on a common physical network infrastructure. It is essential to guarantee quality of services (QoS) by automatically selecting appropriate network and computer resources to virtual networks (consisting of edge, core networks and data-centers) requiring high level of QoS when the surrounding network environments are rapidly-varying. In this paper, we propose to apply a classification method of supervised learning to determining a policy of virtualized resource selection for virtual networks (VNs). When a new VN is constructed, virtualized resources are selected according to the resource selection policy decided by the learning method which derives a classification boundary by means of past records on combinations of service requirements, time-varying network environments and actually-allocated virtualized resources. Then, our proposed scheme automatically and adaptively updates the resource selection policy by autonomously executing relearning on the basis of urgent contingencies, performances of VNs, and so on. Our proposed scheme will make it easier to exclusively occupy virtualized resources guaranteeing QoS especially for VNs requiring high level of QoS by reducing the blocking probability in resources request for the construction of the VNs.

We have proposed the concept of an optical packet and circuit integrated network to provide service diversity, energy efficiency and a simplified control mechanism toward new generation networks. In this integrated network, optical data packets and data on lightpaths are transmitted on common physical resources for efficient resource use. In addition, path signaling for lightpath setup and release thorough optical packet switch block is implemented. We set up a primitive optical packet and circuit integrated network including one switching node and a set of packet/path transceiver. We demonstrate 80 (8λ × 10) Gbit/s colored optical packet switching and 8-lightpaths establishment by transferring optical control packets over the optical packet switching.

Limitations are found in the recent Internet because a lot of functions and protocols are patched to the original suite of layered protocols without considering global optimization. This reveals that end-to-end argument in the original Internet was neither sufficient for the current societal network and nor for a sustainable network of the future. In this position paper, we present design guidelines for a future network, which we call the New Generation Network, which provides the inclusion of diverse human requirements, reliable connection between the real-world and virtual network space, and promotion of social potentiality for human emergence. The guidelines consist of the crystal synthesis, the reality connection, and the sustainable & evolutional guidelines.

In this paper, we propose a chip-level detection and a spectral-slice scheme for the tunable-transmitter/fixed-receiver (TTFR)-based asynchronous spectral phase-encoded optical code-division multiple-access (CDMA) system combined with time-encoding. The chip-level detection can enhance the tolerance of multiple access interference (MAI) because the channel collision does not occur as long as there is at least one weighted position without MAI. Moreover, the spectral-slice scheme can reduce the interference probability because the MAI with the different frequency has no adverse effects on the channel collision rate. As a result, these techniques mitigate channel collisions. We analyze the channel collision rate theoretically, and show that the proposed system can achieve a lower channel collision rate in comparison to both conventional systems with and without the time-encoding method.

In this paper, we propose a mitigation system of high-level multiple access interference (MAI) for multimedia optical Code-Division Multiple-Access (CDMA) systems using the optical power selector (OPS). The proposed system can eliminate high-intensity MAI at the receiver for low-priority users. Moreover, the proposed system can reduce by half the required number of code sequences compared to the conventional scheme. As a result, the proposed system can increase the number of weights at the same code-length and, thus, obtain higher code spreading gain. We analyze performances of the proposed system and show that both high-priority users and low-priority users achieve lower bit error rates in comparison to the conventional scheme.

We have been developing an optical packet and circuit integrated (OPCI) network, which realizes dynamic optical path, high-density packet multiplexing, and flexible wavelength resource allocation. In the OPCI networks, a best-effort service and a QoS-guaranteed service are provided by employing optical packet switching (OPS) and optical circuit switching (OCS) respectively, and users can select these services. Different wavelength resources are assigned for OPS and OCS links, and the amount of their wavelength resources are dynamically changed in accordance with the service usage conditions. To apply OPCI networks into wide-area (core/metro) networks, we have developed an OPCI node with a distributed control mechanism. Moreover, our OPCI node works with a centralized control mechanism as well as a distributed one. It is therefore possible to realize SDN-based OPCI networks, where resource requests and a centralized configuration are carried out. In this paper, we show our SDN architecture for an OPS system that configures mapping tables between IP addresses and optical packet addresses and switching tables according to the requests from multiple users via a web interface. While OpenFlow-based centralized control protocol is coming into widespread use especially for single-administrative, small-area (LAN/data-center) networks. Here, we also show an interworking mechanism between OpenFlow-based networks (OFNs) and the OPCI network for constructing a wide-area network, and a control method of wavelength resource selection to automatically transfer diversified flows from OFNs to the OPCI network.

We propose spectral phase-encoded optical atmospheric (OA) code-division multiple-access (CDMA) systems with pulse-position modulation (PPM) and theoretically analyze the bit error rates (BERs). We compare the BERs of the proposed system to those of time-encoded OA-PPM-CDMA systems at the same bit rate and chip rate. We show that, even if the proposed spectral phase-encoded OA-PPM-CDMA system is equipped with no mitigation techniques of multiple access interference (MAI) and scintillation, the proposed system achieves lower BERs than both time-encoded OA-PPM-CDMA systems, both with and without a mitigation technique of MAI/scintillation, due to its higher code spreading gain and higher tolerance to noise.

We have been developing an optical packet and circuit integrated network (OPCInet) to provide diverse services on the same metro/core fiber network. In this paper, we propose and develop a control method to realize automatic and programmable data transfers from OpenFlow networks (OFNs) to the OPCInet. An OpenFlow controller (OFC) determines how and where to transfer each flow on the OPCInet (i.e. resource selection) on the basis of the information of users' quality-of- service (QoS) requirements, performances in the OPCInet and network traffic in the OFNs. For this, we define QoS mapping between type-of-service (ToS) and ITU-T Y.1541 service classes so that the OPCInet can provide performances satisfying each user's QoS requirement. The proposed method enables effective utilization of wavelength resources with satisfying users' QoS requirements. We implement the automatic wavelength resource selection control in the OFC. We demonstrate that the OFC selects wavelength resources successfully in accordance with our policy of wavelength resource selection.

We propose a novel wavelength resource reconfiguration system for a dynamic and agile new-generation optical switching network such as a high-speed elastic optical network. The system can remove the wavelength channels in time of 1 / (the number of the channels multiplied by 3) compared with conventional method. A proof of concept employing flexible-grid wavelength selective switches and our unique technology burst-mode erbium-doped fiber amplifier demonstration achieved 1 / 12 time in case of removing four-wavelength channels with the high-quality data transmission.

The multi-granular optical-path architecture is a viable approach to creating large-scale optical metro and core networks. We conduct transmission and distributed signaling experiments that emulate the proposed architecture and confirm its technical feasibility.

We design and develop architecture against hardware failures arisen on an optical packet and circuit integrated (OPCI) network. The control plane regularly collects hardware failures' information from the data plane, automatically notifies the failures to the management plane (M-plane), and recovers the status of path usage. For the M-plane, we develop and install a management system based on a hierarchical management model constructed as a relational database to visually manage various OPCI network elements and their correlative relationships over multiple layers. The M-plane contributes to faster discovery of resources affected by hardware failures than the existing OPCI network without any management systems. In verification experiments, we show that hardware failures and resources affected by the failures are successfully displayed on the management window. Besides, the empirically-estimated time to detect a hardware failure and specify affected resources indicates indispensability of the M-plane in that the network administrator can work on recovery operations much faster and earlier and enhance availability of the OPCI network.

In this paper, an M-ary pulse-amplitude modulation (PAM) coded optical code division multiple access (CDMA) system applying the chip-level receiver with M-level threshold detection is proposed. First, in order to increase the number of transmitted bits per pulse, the PAM coded system is considered. However, when the correlator is applied in the receiver, the multiple access interferences (MAIs) with high intensities deteriorate the system performance significantly even if the number of MAIs is small. Consequently, the chip-level receiver with M-level threshold detection instead of the correlator in the M-ary PAM-CDMA system is proposed. The proposed system can obtain the synergistic effect between the PAM and the chip-level receiver, in which the chip-level receiver reduces more MAIs compared with the correlator and the PAM has the larger number of transmitted bits per frame than ON-OFF keying (OOK). As a result, the proposed system can increase the code length and the number of weights, and achieves higher tolerance to MAIs than the OOK-CDMA system with the chip-level receiver under the condition in which the chip duration, the bit rate, and the number of subscribers are kept constant. It is shown that the proposed M-ary PAM-CDMA system with the chip-level receiver achieves better bit error probability (BEP) than the OOK-CDMA system with the chip-level receiver.

We propose a transmission scheme designed to increase the number of subscribers without degradation of the bit error probability in an embedded transmission method for optical code division multiple access (OCDMA) with pulse position modulation signaling (PPM). In the conventional scheme, the transmission symbols and their slot positions are embedded in the desired user's weighted positions in M-ary PPM-OCDMA. However, in the proposed scheme, only the transmission symbols are embedded in the desired user's weighted positions. We analyze the bit error rate (BER) of the proposed scheme for a Poisson photon counting process. We show that the proposed scheme can increase the number of simultaneous users compared to the conventional scheme while achieving a BER of ≤10−9 and keeping the information rate constant. © 2006 Wiley Periodicals, Inc. Electron Comm Jpn Pt 1, 89(10): 55–65, 2006; Published online in Wiley InterScience (www.interscience.wiley.com). DOI 10.1002/ecja.20224

In this paper, we propose a Multibits/Sequence Period Optical Code-Division Multiple-Access (MS-OCDMA) system applying bit erasure correction by differential encoding. In the proposed system, the present transmitted shifted code pattern is determined on the basis of the previous and the present transmitted bits. In the receiver, threshold detection is applied to the avalanche photodiode (APD) output in each shifted code pattern. When the number of APD outputs exceeding the threshold level is either more than two or zero, the present received bit is judged to be an erasure. Then, the present received bit is estimated by the information about the shifted code patterns for which the APD outputs exceed the threshold and the nonerasure bits before or after the present erasure bit, and the erasure bit is corrected. In addition, the proposed system can correct a received bit sequence as an erasure bit instead of an error bit. We analyze the bit error rate (BER) of the proposed system and show that the proposed system has a better BER than the MS-OCDMA system with Reed-Solomon (RS) codes at the same bit rate and with the same number of subscribers. © 2005 Wiley Periodicals, Inc. Electron Comm Jpn Pt 1, 89(2): 1–11, 2006; Published online in Wiley InterScience (www.interscience.wiley.com). DOI 10.1002/ecja.20252

Optical packet switch systems with fiber-delay-line (FDL) buffers provide high-throughput, 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 is not improved drastically in spite of adding extra FDLs. Then, we propose optical and electronic combined buffer architecture. The combined buffer is composed of the 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 the 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 70% larger throughput than that of the existing FDL buffer.

We present an optical packet and circuit integrated (OPCI) ring network which can move a boundary between wavelength resources of optical packets and paths for flexible resource usage. We experimentally demonstrate automatic boundary control between optical packet switch (OPS) and optical circuit switch (OCS) resources using a hybrid optical packet/circuit switch and transponder.

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.

We demonstrated an 8-fiber-delay-line optical buffer with SOA switches for forwarding 100Gbit/s IP-packet-encapsulated optical packets. We derived the maximum number of node hops under error-free operation when buffer scale is increased up to 32 FDLs.

We propose a mechanism for achieving interworking between an optical packet and circuit integrated (OPCI) network and OpenFlow-based networks. Due to the ability of the OPCI network to provide both optical packet switching (OPS) and optical circuit switching (OCS) on the same fiber infrastructure, flows requiring high quality of service can be transferred on lightpaths, while other flows are transferred on large-capacity OPS links. We implemented the interworking function on an OPCI multi-ring network environment and an OpenFlow-enabled network-service testbed. In experiments on Internet Control Message Protocol (ICMP) sending/receiving between end-hosts, we achieved successful interworking between the OPCI network controls and OpenFlow.

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.

We present an SDM-WDM hybrid add-drop network node compatible with SDM-based self-homodyne transmission. OSNR penalties were less than 0.7 dB, but net spectral efficiency was slightly sacrificed.

We develop a common path control-plane for interoperating hybrid optical packet- and circuit-switched ring networks and wavelength-switched optical networks, and experimentally show transparent end-to-end lightpaths can be dynamically established over the two types of networks.

Increasing packet-communication bandwidth is promising for recovery from congestion in large-scale disaster. A forced resource allocation change from packet to circuit in optical packet and circuit integrated networks is demonstrated experimentally.