To achieve the ultra-low latency required by Beyond 5G, realizing direct wavelength connections between users is necessary. Metro networks will also need to accommodate massive multiple wavelengths, several tens of times greater than current wavelength-division multiplexing technology. In this paper, we propose a metro network architecture that achieves multi-band massive multiple wavelengths by using newly developed programmable wavelength selective switches (WSSs). The proposed network introduces a new paradigm of network design and control using Digital Twins. Furthermore, we investigate the impact of remotely located transceivers on the performance of metro networks with programmable WSSs for user direct wavelength connections. We quantify the resource efficiency of the remotely located transceiver model to demonstrate the feasibility of user direct wavelength connection.

This paper proposes a priority-aware guaranteed hardware resource allocation in virtual packet optical nodes (VPONs) and describes experimental evidence of service provisioning with the proposed method on testbed. A network based on the VPON brings solution of service diversification and traffic explosion because it provides multiple services by hardware level separated virtual networks such as IP/Ethernet/ Multi-Protocol Label Switching. The VPON has reconfigurable hardware which is dedicatedly allocated for each service and service-specific logical functions are implemented respectively. However, a previous hardware resource allocation method is not efficient. In previous work, the hardware resources once allocated to a service are not reallocated until the service terminated whenever the hardware is not optimally used. In recent years, hardware reconfiguration time has become shorter and it has enabled to apply for service demand change by releasing hardware from services with low demand and reallocating it to other services with high demand. Thus, this paper proposes preemptive dynamic hardware resource reallocation algorithm for VPON and a priority-aware guaranteed hardware resource reallocation method to realize efficient resource usage and high service capacity. Results of the computer simulation show improving service capability. Furthermore, to demonstrate the feasibility of service provisioning with the proposed method, emulators of the virtual packet optical node were constructed and performed resource reallocation. Results of the experiment show that the virtual packet optical node can configure function chains for providing services based on calculation results of the proposed resource reallocation method.

The lightpath provisioning time or releasing time is adversely affected by the time that optical amplifiers require to adjust to a newly added or terminated signal power. This shortcoming is particularly true with multi-core erbium-doped amplifiers (EDFAs), as multi-core transient-suppressed EDFAs are unavailable at the current time. This paper proposes a fiber span power management scheme based on dummy wavelength signals that are used to shorten the lightpath provisioning and releasing times in multi-core fiber networks. With the shorter time of lightpath provisioning and releasing procedures, the total time that is required to reserve wavelengths in the system is decreased, which means that network resources are used more efficiently. As a result, the blocking performance and average waiting time in the system are improved. To evaluate the performance of the proposed scheme, this paper introduces both analytical model and simulation study. In the introduced model, the ratio of the number of activating and activated dummy wavelengths to the number of dummy wavelengths in each span is considered in the range between 0 and 1. The analysis reveals that the performance of the proposed scheme depends on α, which is the ratio of the number of dummy wavelengths to the number of dummy and lightpath wavelengths in each span, and there exists a point of α where the blocking probability becomes minimum. We further observe that the proposed scheme outperforms the conventional approaches in terms of blocking probability and average waiting time, as traffic loads increase. Finally, we provide the direction on how our introduced model can be considered for a network with multi-span routes.

The continuously growing online shopping is increasing the number of attended home deliveries. The last-mile delivery plays an important role in online shopping satisfaction and especially for food deliveries. This paper focuses on food delivery retailers and particularly investigates the possibility to enhance deliveries using information and data knowledge. In fact, in addition to optimize and to share delivery routes, delivery vehicles could be monitored in order to always maintain shortest delivery delays. We propose in this paper a delivery management architecture targeting these principles. This system is composed of several core mechanisms that should keep delivery delays to a minimum while maintaining low service times. A proof-of-concept of this delivery management system has been developed using Electric Scooters, smartphones and several algorithms. It demonstrates how this architecture could work in a food delivery scenario.

WAN provider links are often operated at low utilization levels, which leaves large unused capacity (headroom). In this paper, we propose using Software Defined Networking (SDN) controllers to support novel Static Headroom (SH) and Dynamic Headroom (DH) services to allow customers to fill this headroom with Elephant Flows (EFs) without adversely affecting the provider's ability to meet its Best-Effort (BE) service-level agreements, and ability to absorb extra traffic load created during failure recovery periods. Our solution calls for the use of lower-priority service for EFs. We use simulations to compare SH service with BE service, and DH service with SH service. When EFs are sent on BE service, they could cause packet losses in general-purpose IP traffic, especially when the burstiness of the latter is high, while with SH service, this packet loss rate is reduced to 0. While DH service requires the added complexity of a provider SDN controller, the ability to dynamically route EFs on lower-utilized links results in higher average EF throughput. The higher the non-uniformity (from a node-pair perspective) in network traffic, the greater the DH gain factor.

Optical networks are well suited to support massive data exchanges between data centers. Elephant traffic flows can be routed over provisioned and dedicated lightpaths (optical flows) while other (mice) flows, which are routed by electronic switches, are unaffected. In some solutions, Wavelength-Selective Switches (WSSs) are employed in the optical nodes to individually route the lightpath towards its destination. WSSs take time to be switched and delay the lightpath setup time. In this paper, the authors compare three service policies for WSS devices aiming to reduce the lightpath setup and tear down times. The conventional service policy assumes that each setup (tear down) request is handled individually. Two other service policies assume that groups of setup (tear down) requests are handled together by the WSS. These policies are implemented in a discrete event simulator, which is used to estimate the end-to-end lightpath setup and tear down time across an arbitrary mesh network. Simulation results show that group service policies outperform the conventional policy at high loads. The grouping policies are useful to reduce the lightpath setup time especially in the presence of lightpaths that are frequently set up and have relatively short holding time (short duration of elephant-optical flows).

Recently, the energy consumption in data centers is increasing. Introducing optical circuit to the inter-rack communication is considered as an effective solution. The authors are working on the research and development of the High-speed Optical Layer 1 Switch system for Time slot switching based optical data center networks (HOLST). We focused on Plomb Lanthanum Zirconate Titanate (PLZT) optical switch that performs nano seconds switching time. In this paper, several ideas on how to connect PLZT optical switches and Micro Electro Mechanical Systems (MEMS) optical switches, the composing modules of the HOLST, are proposed to design network architecture and evaluated in order to compare and consider power saving effect. The result of the numerical simulation shows that the proposal architecture achieves 50% energy consumption reduction than the conventional architecture.

In a high-speed backbone network, the failure of a network link may cause large data losses, so it is necessary to reserve spare network resources for faster recovery. The conventional protection methods to reserve backup routes do not consider the failure probability of each network link, so the same amount of network resources for the backup route are needed regardless of the failure probability of network links. This leads a decrease in the number of connections that can be accepted into the network. This paper proposes a routing and capacity allocation method that guarantees the expected value of allocated capacity. We formulate a mixed integer liner programming model for the proposed method. We conduct simulations to study the advantage of the expected capacity guaranteed routing over the conventional routing method in terms of bandwidth blocking probability. The results show that the proposed method reduces the bandwidth blocking probability to about 1/3 as compared to that of the conventional path protection method.

In the smart grid, which is a new generation grid integrated with bidirectional communication and advanced information technology, it is supposed that the number of energy suppliers will increase because many renewable resources will be connected to the grid. The increase in the number of suppliers that use renewable energy resources introduces new challenges in scheduling and dispatching controllable resources to control the load of the power grid in the system. By using the pricing method that maximizes social welfare, consumers/suppliers can determine the optimal power consumption/generation that maximizes own welfare. However, when a forecast error of renewable energy output occurs, the gap between the total power consumption and generation causes unless consumers decrease power consumption from the optimal consumption. Decreasing consumers' power consumption might lower their utility. In this paper, we define impact on consumers' utility and propose two demand response methods which aim to minimize the impact. The first method, called One-Time Demand Response, minimizes the impact on consumers' utility of a certain time slot by using a Lagrange multiplier. In the second method, called Foresight Demand Response, all consumers can forecast the electricity market price of the next time slot by using linear regression, and consumers do not reduce power consumption when the incentive of a certain time slot is cheaper than the forecast electricity market price of the next time slot. From the performance evaluation, we confirm the impact on consumers' utility in One-Time Demand Response and Foresight Demand Response. We also confirm that both two proposals can increase consumers' welfare compared to not carrying out demand response.

In Smart Grid, various kind of renewable energies have been rapidly introduced. Then, the structure of power grid has been shifting from the tree 1:N structure with centralized power plants to the M:N liked mesh structure with various kind of power resources. In this case, it is necessary that each power demand user determines to use a proper power source among a huge number of power supplies, and we call it as “Matching” in this paper. There are three issues to realize M:N matching control. The first issue is reducing calculation time. The second issue is the dynamic adjusting in spite of fluctuation of power demand and supply. The third issue is to control user satisfied matching such as power price or satisfaction to select renewable energy. In this paper, we propose the matching method that the demand and supply matching algorithm which solves above three issues by utilizing the P2P Interactive Agent eXtensions (PIAX) platform. The calculation time can be reduced by the autonomous distributed control by PIAX platform. The issues of dynamic adjusting of power demand and supply and to consider user satisfaction are solved by the proposed matching algorithm. Computer simulation results and evaluation results by PIAX platform show that the proposed method can reduce the matching calculation time, and maximize user satisfaction in power fluctuation of demand and supply.

A flexible orthogonal frequency-division multiplexing optical network enables to change bandwidth flexibly by changing the number of sub-carriers. We consider that users request to dynamically change the number of sub-carriers. In this context, the network resources can be used more efficiently. The dynamic bandwidth change needs a certain time. Service centric resource allocation must be considered in terms of waiting time to change the number of sub-carriers. If the user demands drastically increase such as a disaster priority telephone service, a waiting time for bandwidth changes is not tolerated because emergency is time-critical. This is caused by a chain-change of bandwidth such as a multiple-car pileup. This paper proposes a grouped elastic spectrum allocation scheme to satisfy the tolerable waiting time of the service in an optical fiber link. Spectrums are grouped to restrict a waiting time in the proposed scheme. In addition, the proposed scheme determines a margin bandwidth with neighbor spectrums to prevent frequent reallocation by estimating real traffic behavior. Numerical results show that a required bandwidth can be saved with satisfying waiting time constraints. Additionally measurement granularity is discussed.

This paper proposes a source-based high-speed wavelength-path protection scheme configuring a tree-shaped backup path in WDM networks. In the proposed scheme, a failure-detecting node on a primary path starts protection by sending a failure notification to a source node of the primary path through the tree-shaped backup path. Intermediate nodes on the backup path and the source node which receive the failure notification switch to the backup path autonomously. Therefore, the proposed scheme provides high-speed protection. The simulation results revealed that the proposed scheme reduces the failure-recovery time compared with the conventional protection schemes.

Elastic Optical Networks (EONs) have the potential to improve the fiber spectrum utilization by allocating spectrum resources to multiple traffic requests proportionally to the amount of carried traffic. However, achieving high spectrum utilization in this elastic scenario is hindered by the resulting spectrum fragmentation. A number of studies have addressed and made attempts to mitigate spectrum fragmentation. Most of these studies are based on simulation techniques and target the overall blocking probability experienced by the offered traffic requests due to the lack of available spectrum resources. Some studies have also shown that blocking probability in EON can be uneven, i.e., high-rate circuit requests are more likely to be blocked when compared to low-rate requests due to the shortage of contiguously available spectrum resources. The contribution of this paper is to extend an existing Markov Chain (MC) model previously proposed by the authors to quantify blocking probability in a two-service elastic fiber link. The model extension accounts for a self-limited and partial sharing of the fiber spectrum to accommodate the two types of service. The MC model is used to quantify both the blocking probability and its fairness across the two types of service, documenting how the EON uneven blocking behavior can be significantly mitigated by performing partial (as opposed to full) sharing of the fiber spectrum.

This paper aims to design a Hadoop system and evaluates the performance of a task allocation scheme. The task allocation scheme splits each job into tasks using an appropriate splitting ratio, and assigns tasks to slave servers based on server processing performance and network resource availability. We experimentally evaluate the performance of the scale out of the task allocation scheme with five machines. We focus on the configuration of jobtracker and tasktracker in Hadoop. In cases with heterogeneous Hadoop clusters, we distribute task blocks to high-capability slaves with proportionally larger-sized tasks than to low-capability slaves. We create an environment in which high-capability slaves perform more work than low-capability slaves. The experimental testbed results indicate that the task allocation scheme is effective.

Software switches offer flexibility to service providers but potentially suffer from low performance. A software switch called Lagopus was implemented using Intel's Data Plane Development Kit (DPDK), which offers libraries for high-performance packet handling. Prior work on software switches focused on characterizing packet forwarding throughput. In this work, we evaluated the impact of certain parameters and settings in Lagopus on application performance and studied packet drop rates. The importance of receive-thread packet classification for load balancing and to send delay-sensitive flows to a different worker thread from high-throughput flows was first demonstrated. Next, we showed that a loop-count variable used to control packet batching should be kept small in case link utilization is low. Finally, we showed that packet drop rate could be non-zero when the OpenFlow table size is large and packet arrival rate is high, and interestingly, the packet drop rate was higher with four worker threads than with a single worker thread. This implies a need for careful calibration and planning of the parameters of parallelization.

The recent development of network technologies that offer centralized control of explicit routes opens the door to the online optimization of explicit routing. For this kind of Traffic Engineering optimization, raising the calculation speeds by using multi-core processors with effective parallel algorithms is a key goal. This paper proposes an effective parallel algorithm for General purpose Programming on Graphic Processing Unit (GPGPU); its massively parallel style promises strong acceleration of calculation speed. The proposed algorithm parallelizes not only the search method of the Genetic Algorithm, but also its fitness functions, which calculate the network congestion ratio, so as to fully utilize the power of modern GPGPUs. Concurrently, each execution is designed for thread-block execution on the GPU with consideration of thread occupancy, local resources, and SIMT execution to maximize GPU performance. Evaluations show that the proposed algorithm offers, on average, a nine fold speedup compared to the conventional CPU approach.

Elastic Optical Networks (EONs) enable optical circuits to be assigned distinct numbers of spectrum slices. Individual circuits can then be assigned an optimal number of slices to best match their target transmission rates. A well-known drawback of EONs is spectrum fragmentation and its resulting uneven blocking probability, which circuit requests experience when the available spectrum slices in the fiber are insufficient or not contiguous. Capturing this spectrum fragmentation problem analytically is a challenging problem. Not surprisingly, most of the existing studies at this time mainly use simulation-based techniques to quantify blocking probability in EONs. In this paper, the authors present a Markov Chain (MC) model that attempts to characterize the fragmentation problem in a simplified scenario, i.e., only two types of circuit services are allowed over a single fiber link. Despite its limited scope, this initial analytical effort is able to accurately capture the non-monotonic behavior of the blocking probability in EONs for the first time.

This paper proposes a new two-wavelength design for large-enterprise access links. The goal of this design is to lower power consumption and equipment costs without having a significant impact on performance. In our design, one wavelength is used as part of a lower-rate static circuit for general-purpose IP traffic, while the second wavelength is dynamically configured into a high-rate circuit for large dataset transfers whenever needed. A few provider-router ports are shared among a larger number of customers given that large dataset transfers are relatively infrequent. This leads to potential start-time delays, but results in significant power and cost savings. Using analytical models, we provide a quantitative power-and-cost comparison of our dynamic solution with the static always- on high-rate single-circuit solution used currently. The number of shared provider-router ports is kept large enough based on the number of customers to ensure that the probability of start-time delay exceeding a threshold is less than 1%.

Information-Centric Networking (ICN) employs communication based on location independent data identifiers (IDs) that enables efficient data dissemination. However, some important challenges still remains in the conventional ICN schemes when they are applied to a highly mobile and distributed environment, such as growth in amount of route information or increased transaction load of name resolution services. To accommodate this, we propose a new ICN scheme where routes are aggregated at aggregation nodes and optimized routes are established during data registration transactions to mitigate the amount of route information, delay, and transaction load. We evaluate the proposed scheme using a nation-wide test bed of 37 intermediate nodes and 336 mobile terminals. We show that the proposed scheme enables efficient data dissemination where the mobile terminals register and retrieve data in widely distributed manner.

This article proposes a network control method that realizes on-demand service providing in Elastic Lambda Aggregation Network (EλAN). EλAN is a next-generation access/aggregation network that provides various network services having different protocols and QoS requirements. EλAN also achieves the reduction of total network power consumption by sharing access network equipment by multiple network services. In the proposed method, fixed optical trees are introduced to EλAN to exchange control messages in the same manner as today's FTTH systems. Discovery process and optical path/tree aggregation using the optical trees are explained in this article. We also discuss the energy-saving effect of the optical path/tree aggregation in EλAN.

This paper presents a concept of the Energy Efficient and Enhanced-type Data-centric Network (E3-DCN). For future network technology, the Contents-centric Network (CCN) is widely studied. DCN is an extended version of CCN. M2M communication and mobility of the data are supported. In the enhanced-type DCN, a user requests data name or identifier to the network. If the exactly matched requested data are not found in the network, the network constructs the requested data. E3-DCN provides an energy efficient contents delivery network on the network virtualization platform.

We propose a new efficient singlecast / multicast method for active optical access network using PLZT 10 nsec high-speed optical switches. The Active Optical Network, called ActiON, has been proposed using slot type switched optical network. Compared with Passive Optical Network (PON), ActiON can quadruplicate the number of subscribers (128 users) per OLT and double the maximum transmission distance (40 km) between OLT and ONUs. However, ActiON uses slot based switching method, so it is difficult to realize the multicast delivery. In this paper, we propose the efficient singlecast / multicast method for ActiON by using PLZT optical switch elements which are controlled as “distribution mode” like an optical splitter by applying mid-control voltage. In addition, a new efficient multicast slot allocation method is proposed and formulated as a linear programming problem. This formula develops the maximum number of users which is able to be connected by multicast and the minimum number of slots is used for multicast users.

This paper proposes a fast fault recovery method for the wide area Ethernet. To achieve fast recovery, a pre-established forwarding table is introduced. The pre-established table is looked up when a fault occurs, and frames are forwarded without frame loss. Ethernet lacks of fault recovery features required in Wide Area Network (WAN) since it is originated as Local Area Network (LAN) technology. Protection or restoration is a typical method for fault recovery, but a lot of frame losses are inevitable. Our proposal can decrease frame losses since frames are forwarded without waiting for path switching to be completed to use the pre-preparation table. In addition, the proposed method can be operated with the legacy forwarding method seamlessly. The simulation results show that frame loss decreases to 40% or less with our proposed method.

This paper proposes a parallel shortest path-searching algorithm and implements it on a newly structured parallel reconfigurable processor, DAPDNA-2 (IPFlex Inc). Routing determines the shortest paths from the source to the ultimate destination through intermediate nodes. In open shortest path first (OSPF), Dijkstra's shortest path algorithm, which is the conventional one, finds the shortest paths from the source on a program counter-based processor. The calculation time for Dijkstra's algorithm is O(N2) when the number of nodes is N. When the network scale is large, calculation time required by Dijkstra's algorithm increases rapidly. It's very difficult to compute Dijkstra's algorithm in parallel because of the need for previous calculation results, so Dijkstra's algorithm is unsuitable for parallel processors. Our proposed scheme finds the shortest paths using a simultaneous multi-path search method. In contrast with Dijkstra's algorithm, several nodes can be determined at one time. Moreover, we partition the network into different groups (network groups) and find the all-node pair's shortest path in each group using a pipeline operation. Networks can be abstracted, and the shortest paths in very large networks can be found easily. The proposed scheme can decrease calculation time from O(N2) to O(N) using a pipeline operation on DAPDNA-2. Our simulations show that the proposed algorithm uses 99.6% less calculation time than Dijkstra's algorithm. The proposed algorithm can be applied to the very large Internet network designs of the future.

A deadline-aware scheduling scheme for the lambda grid system is proposed to support a huge computer grid system based on an advanced photonic network technology. The assignment of wavelengths to jobs in order to efficiently carry various services is critical in lambda grid networks. Such services have different requirements such as the job completion deadlines and wavelength assignment must consider the job deadlines. The conventional job scheduling approach assigns a lot of time-slots to a call within a short period in order to finish the job as quickly as possible. This raises the blocking probability of short deadline calls. Our proposal assigns wavelengths in lambda grid networks so as to meet QoS (quality of service) guarantees. The proposed scheme assigns time-slots to a call over time according to its deadline, which allows it to increase the system performance in handling short deadline calls, for example, lowering their blocking probability. Computer simulations show that the proposed scheme can reduce the blocking probability by a factor of 100 compared with the conventional scheme under the low load condition in which the ratio of long deadline calls is high. The proposed scheduling scheme can realize more efficient lambda grid networks.

In this paper, we propose a new wavelength assignment scheme that improves the blocking probability of WDM networks that use limited-range wavelength converters. Limited-range wavelength converters are attractive for wavelength-routed networks given current technology since they offer good utilization of the wavelength resource and improved blocking probability. However, their conversion ranges are limited, the maximum difference between the input and output wavelengths is restricted. Thus, we must take into account the existence of these limited-range wavelength converters. In our proposed scheme, each connection request is assigned a different wavelength according to its hop number. We tend to use different wavelengths for connection requests with different hop numbers. As a result, we can reduce the blocking probability by two decades compared to simply assigning the shortest available wavelengths. In addition, it allows the number of wavelength converters used in each node to be reduced with almost no degradation in blocking probability. Simulation results show that the proposed scheme can reduce the wavelength converters by about 20%.

Although the Internet is playing an increasingly significant role in global communication, it remains vulnerable to malicious traffic such as worms and DoS/DDoS attacks. In the last few years, the emergence of high speed active worms, such as Code Red II, Nimda, SQL Slammer and MS Blaster, has become a serious issue. These worms cause serious damage to communication networks throughout the world by using up network bandwidth. In addition, since conventional firewall systems are located just in front of the server and do not prevent malicious traffic from entering the network, they cannot prevent such network congestion. We have developed a proto-type system of a network firewall using reconfigurable processors. In this paper, we overview the developed system and present its evaluation results.

The paper proposes the hierarchical cloud-router network (HCRN) to solve the problem of overcoming the scalability limit in a multi-layer generalized multi-protocol label switching (GMPLS) network. We define a group of nodes as a virtual node, called a cloud-router (CR). A CR consists of some number of nodes or lower-level CRs. A CR is modeled as a multiple switching capability (SC) node when it includes more than one kind of SC, such as fiber SC, lambda SC, time-division multiplexing (TDM) SC, packet SC, even if there are no actual multiple-SC nodes in the CR. The CR advertises its abstracted CR internal structure which is abstracted link state information inside the CR. A large-scale, multi-layer network can then achieve scalability by advertising the CR internal structure throughout the whole network. In this scheme, the ends of a link connecting two CRs are defined as interfaces of the CRs. We adopt the CR internal cost scheme between a CR's interfaces to abstract the network. This CR internal cost is advertised outside the CRs via the interfaces. Our performance evaluation has shown that HCRN can operate a larger number of nodes than a normal GMPLS network. It can also bear more frequent network topology changes than a normal GMPLS network.

This paper proposes a distributed virtual network topology (VNT) reconfiguration method for Internet Protocol over a wavelength-division-multiplexing network under dynamic traffic demand. We have developed a simple heuristic algorithm for calculating the VNT for distributed control. A generalized multiprotocol label switching (GMPLS)-based routing protocol has been developed. The VNT is quickly reconfigured by setting up and/or tearing down lightpaths using a GMPLS signaling protocol. Traffic demand is measured at the ingress node and advertised by the extended GMPLS routing protocol. Performance of the proposed method is investigated using variable traffic model.

The paper describes multi-layer traffic engineering and signaling technologies in photonic-GMPLS-router networks. Multi-layer traffic engineering, which yields the dynamic cooperation of IP and photonic layers for providing IP services cost-effectively, is described. To realize multi-layer traffic engineering, we propose an OSPF extension that advertises both the number of total wavelengths and the number of unused wavelengths and an RSVP-TE extension that minimizes the number of wavelength conversions needed. In addition, the paper presents a heuristic-based multi-layer topology design scheme that uses IP traffic measurements in a generalized multiprotocol label switch (GMPLS). Our design scheme yields the optical label switch path (OLSP) network topology, i.e. OLSP placement, that minimizes network cost, in response to fluctuations in IP traffic demand. In other words, the OLSP network topology is dynamically reconfigured to match IP traffic demand. Networks are reconfigured by the proposed scheme so as to utilize the network resources in the most cost-effective manner.

Many address lookup methods on IP routers have been recently proposed to improve the packet forwarding capability; nevertheless, their performance prediction is very limited because of lack of consideration of actual traffic characteristics in their evaluations. It is necessary to consider actual traffic to predict more realistic performance in routers, especially in the case of layer 3 and 4 switches whose performance is more influenced by flow characteristics. In this paper, we propose new methods for predicting the router's performance based on the statistical analysis of the Internet traffic. We also present an example of its application to the existing table lookup algorithm, and show that simulation results based on our method can provide accurate performance prediction.

We propose a buffer management mechanism, called V-WFQ (virtual weighted fair queueing), for achieving approximately fair bandwidth allocation with a small amount of hardware in high-speed networks. The basic process for allocating bandwidth fairly uses selective packet dropping to compare the measured input rate of the flow with an estimated fair bandwidth share. Though V-WFQ is a hardware-efficient FIFO-based algorithm, it can achieve almost ideal fairness in bandwidth allocation. Simulation results show that V-WFQ achieves a good balance between fairness and link utilization under various simulation conditions.

Next-generation Internet backbone node architectures are proposed that offer sophisticated QoS support including real-time voice, video streaming, and VPN. The backbone core node, named MSN ( multi-service network) Type-X, has 320 Gbit/s throughput and IPv4/v6 dual stack forwarding capability. Note that, through the employment of state-of-the art VLSI technology, both v4 and v6, OC-192C full wire-rate forwarding performance is achieved for the first time. In addition, sophisticated DiffServ with programmable thousands of virtual queues is also proposed to realize a complete set of hardware. In addition, Type-X uses an ATM emulation technique that can into allow the new packet-centric backbone to support conventional services. Finally, the paper describes high-reliability techniques such as newly structured service non-stop software and a firmware upgrade technique. These newly proposed techniques and systems will realize the next-generation, real-time, high-QoS capability, and VPN IP backbone network.

This paper describes a wavelength-division-multiplexing (WDM) optical switching technique that achieves 5-Tb/s switch throughput. This technique is employed in our previously described OPTically Interconnected Distributed Multi-stage Tb/s-ATM switching Network Architecture, called OPTIMA-2. This is a scalable and non-blocking 3-stage switch employing optical WDM and dynamic bandwidth sharing. An inter-stage, cell-based WDM switch is fabricated in combination with field programmable gate arrays (FPGAs) and 2.5-Gb/s 8-wavelength optical interconnection modules. Experimental results from a 5-Tb/s switching system are introduced.

A new scheduling algorithm that concatenates data blocks to increase switching throughput is proposed. The algorithm controls the degree of concatenation and reduces the number of switching instances to improve switch utilization. The switch architecture uses virtual output queue switching architecture where the core switch fabric is an optical matrix switch. The optical matrix switch requires the guard-time overhead needed for optical switch control. By reducing the number of switching instances, guard-time overhead can be reduced and switch utilization can be improved. Computer simulations show that efficiency is dramatically increased and that fairness in terms of data throughput among output ports is achieved.

The hardware architecture of a 5-Tbit/s FB (frame-based) ATM switching system OPTIMA-2 (OPTically Interconnected Distributed Multi-stage Tbit/s-ATM switching Network Architecture-2) is described. OPTIMA-2 has a non-blocking 3-stage switch architecture employing optical WDM links and dynamic bandwidth sharing. The WDM links are composed of a sender port, wavelength arrayed waveguided grating (AWG) router and receiver port. The sender port at the output port of a switch-element allocates a packet to one of eight WDM wavelengths and transmits it. The receiver port receives packets of all eight wavelengths via the wavelength AWG router and merges them at an input port of the next-stage switch-element. Total bandwidth is limited to 10 Gbit/s, but the maximum bandwidth of each wavelength is designed to be 2.5 Gbit/s to prevent the statistical multiplexing gain from falling. A scheduler, which selects variable-length packets from the eight wavelengths, can keep fairness and a small delay. The bandwidth of each wavelength is changed dynamically by the system so that the traffic in each wavelength is equally distributed among the total bandwidth. In OPTIMA-2, variable-length FB-ATM cell, which is familiar with the IP packet, can be switched while keeping the throughput fairness.

This paper proposes a 3-stage ATM switch architecture that uses optical WDM (wavelength division multiplexing) grouped links and dynamic bandwidth sharing. The proposed architecture has two features. The first is the use of WDM technology which makes the number of cables used in the system proportional to system size. The second is the use of dynamic bandwidth sharing among WDM grouped links. This prevents the statistical multiplexing gain offered by WDM from falling even if the switching system becomes large. A performance evaluation confirms the scaleability and cost-effectiveness of the proposed architecture.

An experimental 640-Gbit/s ATM switching system is described. The switching system is scalable and quasi-nonblocking. It uses hardware self-rearrangement in a three-stage network. Hardware implementation results for the switching system is presented. The switching system is fabricated using advanced 0.25 /spl mu/m CMOS devices, high-density multi-chip-module (MCM) technologies, and optical wavelength division multiplexed (WDM) interconnection technologies. A scalable 80-Gbit/s switching module is fabricated combined with a newly developed scalable-distributed-arbitration technique, and a WDM interconnection system that connects all 80-Gbit/s switching modules is developed. Using these components: an experimental 640-Gbit/s switching system is partially constructed. It can be applied to realize future broadband ATM networks.

This paper proposes a new high-speed network architecture DBTN (dynamic burst transfer time-slot-base network), which is based on circuit-switched network technology. A routing tag is attached to a burst at an ingress edge node and the burst is self-routed in a DBTN network by creating a circuit dynamically. The routing tag, which is called time-slots-relay, is a series of link identifiers to the destination. The time-slots-relay consists of link identifiers from the ingress to the egress nodes and is used to create the circuit. Subsequent data is switched over the circuit being created in an on-the-fly fashion. A link identifier in the time-slots-relay is unique within each node. Each link identifier is loaded into address control memory (ACM) of each circuit-switched transit node, and the circuit to the destination is established. Subsequent user data follows immediately after the time-slots-relay and is sent over the established circuit. Thus short-lived fairly large data transfers like WWW traffic are efficiently carried. A circuit between adjacent nodes is created and released dynamically so bandwidth efficiency is improved compared with conventional circuit-switched networks. Time division multiplexing of the circuit-switched network is utilized so there is no delay jitter or loss. We address the performance of DBTN switches and report on the implementation of the system.

A low-pass filter (LPF) for ATM traffic measurement is presented and an admission control method based on it is proposed. Instantaneous rate is estimated with the LPF and the maximum instantaneous rate (MIR) observed over the monitoring period is used to determine the residual bandwidth in the method. Hence we call it the MIR method. We developed an analytical model to investigate the performance of the MIR method. We confirmed that the MIR method regulates the number of connections properly even for connections with long holding time. In addition, the MIR method is robust against long range dependent traffic, which degrades the robustness of many previously proposed methods. Because the MIR method does not assume any mathematical model and can be implemented with simple hardware, it is well suited for practical ATM switching systems.

Proposes active-ATM, a flexible, simple and cost-effective ATM-WAN architecture that can handle multiple user-customized ATM-layer protocols, such as ABR and ABT, by using a simple universal ATM transit network. The architecture enables the construction of flexible networks that can evolve easily. With active-ATM and the ATM multi-protocol emulation network architecture called ALPEN, it is easy to implement new ATM-layer protocols by using user-created programs called active-program capsules that modify only the edge nodes. Because these user-sent program capsules can be used to quickly customize the edge nodes, there is no waiting for standardization and implementation of new services. The ATM-layer protocols are emulated only at the edge nodes, making the transit network independent of customer ATM-layer protocols. The active-ATM edge node is based on the flexible programmable node architecture called PUN (programmable unified node). PUN is a platform for user-programmable ATM-layer services. A prototype system has demonstrated the flexibility of the resulting ATM network. The simple and high-speed universal ATM transit network periodically reports to the edge nodes the performance of the its routes by using periodical route performance check sequences, which are independent of the ATM-layer protocols. The information is used by the edge nodes to manage user cell rates in real time. Flexible, adaptive, and sophisticated protocols that efficiently utilize network resources can be easily supported by the network's edge nodes.

A new connection admission control method based on actual virtual path traffic measurements is proposed to achieve high bandwidth efficiency for various types of traffic. The proposed method is based on the measurement of instantaneous virtual path utilization, which is defined as the total cell rate of the active virtual channels normalized by the virtual path capacity. A low-pass filter is used to determine the instantaneous virtual path utilization from crude measurements. A smoothing coefficient formula is derived as a function of the peak rate of the virtual channel. The residual bandwidth is derived from the maximum instantaneous utilization observed during a monitoring period. Simulation shows that the proposed method achieves statistical multiplexing gains of up to 80% of the limit possible with optimum control for similar traffic sources. It can be implemented with very simple hardware. The admission decision is simple: the requested bandwidth is compared with the residual bandwidth. This method is therefore well suited for practical asynchronous transfer mode switching systems.

This paper proposes a rate control technique for high-speed burst transmission with quick response in ALPEN, an ATM multi-protocol emulation network. Under lower load, it is desirable for the network to allow very quick burst access to a lot of bandwidth, like UBR. Under higher load, a mechanism for congestion avoidance or recovery is required, like ABR. The rate control technique proposed provides both properties. In ALPEN, the transit switch sends only information about the residual bandwidth to the source side edge switch. The edge switch responds to user requests quickly because it knows route status. This is very effective because long delay in ATM-WAN reduces response performance. The proposed technique improves ATM-WAN performance because it has a algorithm to collect the information about route status without using any knowledge about delay length. Its performance is confirmed by simulations.

A high-speed and distributed ATM switch architecture, called the TORUS switch, is proposed with the aim of achieving a terabit-per-second ATM switching system. The switch is a distributed and scalable internal speed-up crossbar-type ATM switch with cylindrical structure. The self-bit-synchronization technique and optical interconnection technology are combined to achieve gigabit-rate cell transmission, where high-density implementation technologies such as multichip module technology are not required at all. Also, distributed contention control based on the fixed output-precedence scheme is newly adopted. This control is very suitable for high-speed devices because its circuit is achieved with only one gate in each crosspoint. A TORUS switch is fabricated as a 4/spl times/2 switch module using optical interconnection technology and very high-speed crosspoint LSIs, constructed using an advanced Si-bipolar process. Measured results confirm that the TORUS switch can be used to realize an expandable terabit-rate ATM switch.

This paper describes a real-time emulation method based on the use of commercial FPGAs, high-speed interconnect switch devices and software tools (gate array LSI to/from FPGA netlist migration and multi-FPGA partitioning tools) to emulate large scale high-speed communication switching system without the need for conventional specific wiring structures between FPGAs and timing control circuit insertion. This method is applied to line interface units (LUs) that have 20 MHz operation; we employ the LU board and emulation boards, both of which have hierarchical structures with small sub-boards. The proposed LU board structure has the advantage of utilizing a common LU board for rapid prototyping as well as developing cost effective, practical high-performance Asynchronous Transfer Mode (ATM) switching systems for broadband ISDN (Integrated Services Digital Networks). The results of multiple FPGA partitioning on the emulation board suggest that the proposed method will yield economic systems that are also practical.

A broadband network architecture is proposed that integrates multimedia services, such as data, video, and telephony information, using 52-Mb/s based STM-paths at the user network interface (UNI). The user can access any new service via the STM-based access network via either synchronous transfer mode (STM) switching or asynchronous transfer mode (ATM) switching. STM circuit switching supports long duration, constant bandwidth data transfer services such as video and high-definition television (HDTV) distribution and will also be used for the crossconnect system. Circuit switching can provide transparent transmission during long connection periods. This paper also proposes an expandable time-division switch architecture, an expandable time-division switching LSI, and an expandable switching module for small to large size system applications. The proposed time-division switching LSI, module, and system handle 52-Mb/s bearer signals and have throughputs of 2.4 Gb/s, 10 Gb/s, and 40 Gb/s, respectively. The time-division switch realizes video distribution with 1:n connections. Finally, a local switching node that features an expandable 52-Mb/s time-division circuit switching network is shown for multimedia access networking.

The performance limitation of the "leaky bucket algorithm" is analyzed for usage parameter control and traffic management in asynchronous transfer mode (ATM) networks. Simulation results show that the conventional statistical bandwidth allocation method, which uses "the worst pattern derived from the cell interarrival time moments" permitted by the leaky bucket algorithm, can not guarantee the quality of service (QOS). As a result, this paper proves that the VC/VP bandwidth allocation method based on the leaky bucket algorithm is unsatisfactory.>

A gigabit-rate five highway interface GaAs optoelectronic LSI chipset has been fabricated for the 0.85 mu m wavelength range optical interconnections between modules or VLSIs. The optical sender consists of a high-speed laser driver array LSI having 2 Gb/s maximum operation speed and a tiny laser array. The optical receiver in a GsAs high-speed optical receiver array LSI with a monolithically integrated metal-semiconductor-metal (MSM) photodetector, a high-speed preamplifier, and a decision circuit that has a maximum operation speed of 1.8 Gb/s. The receiver LSI is provided with a new bit-synchronizing circuit and an automatic threshold determination circuit. >

An expandable space-division (SD) switch architecture and a bipolar circuit design for gigabit-per-second crosspoint-switch LSIs are described. An expandable 2-Gb/s 16*16 crosspoint switch LSI which employs a novel switch structure, a novel circuit design, and a super self-aligned process (SST-1A) is developed. A switching module and partial 1:n nonblock, full 1:1 nonblock switching network architecture are also presented. Using the LSI and the switching network architecture, an experimental 620-Mb/s network system is demonstrated.>

The authors propose a novel structure for a high-speed multihighway serial-to-parallel converter for future high-speed digital switching systems. The proposed structure consists of a two-dimensional shift register array which can shift both horizontally and vertically. The shift direction is changed at the frame periods. The calculated results confirmed the superiority of the proposed structures over conventional ones, namely, faster operating speed and fewer gates.>

This paper describes the high-speed time division switch employed in a 32-Mbit/s bearer signal communications system. System performance is realized by using three technologies. The first is a switch structure referred to as a 2-RAM 2-bank structure which ensures high-speed performance by increasing switching throughput four times over that of the basic structure. The second is the inclusion in the switch of two types of peripheral logic developed using Si-bipolar super-self-aligned process technology. The third is high-speed synchronous transmission of data. A large channel capacity time division switching network is also discussed. In conjunction with the network, these technologies make it possible to realize the ISDN time division switches necessary for such services as TV and high-definition TV communications.

A novel concept and structure for a high-speed time division switch operating in the gigabit/second range are presented. The basic concept is to use the high-speed memory in the time switch effectively by means of a slow writing/fast reading method.Nnumbers of speech data are written in parallel duringNclock times and read serially during one clock time for one switched datum. By using the concept and high-speed memories that we have, a data switching system which approaches the memory and read cycle limit can be realized with a proper time margin.