Differentiated-services model has been prevailed as a scalable solution to provide quality of service over the Internet. Many researches have been focused on per hop behavior or a single domain behavior to enhance quality of service. Thus, there are still difficulties in providing the end-to-end guaranteed service when the path between sender and receiver includes multiple domains. Furthermore differentiated-services model mainly considers quality of service for traffic aggregates due to the scalability, and the quality of service state may be time varying according to the network conditions in the case of relative service model, which make the problem more challenging to guarantee the end-to-end quality-of-service. In this paper, we study class mapping mechanisms along the path to provide the end-to-end guaranteed quality of service with the minimum networking price over multiple differentiated-services domains. The proposed mechanism includes an effective implementation of relative differentiated-services model, quality of service advertising mechanism and class selecting mechanisms. Finally, the experimental results are provided to show the performance of the proposed algorithm.

While mobile networks have been enhanced to support a variety of mobile multimedia services such as video telephony and rich data content delivery, a new challenge is being created by the remarkable development of micro-device technologies such as micro processor-chips, sensors, and RF tags. These developments suggest the rapid emergence of the ubiquitous computing environment; computers supporting human life without imposing any stress on the users. The combination of broadband global networks and ubiquitous computing environment will lead to an entirely new class of services, which we call ubiquitous networking services. This paper discusses how to create ubiquitous service environments comparing global networking approaches which are based on fixed and mobile networks. It is shown that the mobile approach is better from service applicability and reliability viewpoints. Networking architecture is proposed which expand 4G mobile cellular networks to real space via gateways on the edges of the mobile network (i.e. mobile terminals). A new set of technical requirements will emerge via this approach, which may accelerate the paradigm shift from the current mobile network architecture and even from the Internet of today.

We have developed an experimental approach that allows us to estimate the performance of a large-scale enterprise network to update routing information. This approach was applied to the integration of the UFJ Bank network system on January 15, 2002. The main characteristic of this approach is the application of a formula that represents the delays in updating routing information that accompany reductions in CPU resources. This procedure consists of two steps: one is to estimate the reduction in the availability of CPU resources caused by forwarding of data packets at a router, and the other is to estimate the levels of CPU resources required for replying to a query about a new route and subsequently updating the routing information. These steps were applied to estimate the performance of the network in terms of routing information convergence. The results of our experiments on the network showed that updating the routing information was possible as long as the average level of CPU utilization during any five-minute period at the routers was less than 40%. We were able to apply this guideline and thus confirm the stability of the UFJ Bank network.

A principal attraction of ATM networks, in both wired and wireless realizations, is that the key quality of service (QoS) parameters of every call, including end-to-end delay, jitter, and loss are guaranteed by the network when appropriate cell-level traffic controls are imposed at the user network interface (UNI) on a per call basis, utilizing the peak cell rate (PCR) and the sustainable cell rate (SCR) values for the multimedia--voice, video, and data, traffic sources. There are three practical difficulties with these guarantees. First, while PCR and SCR values are, in general, difficult to obtain for traffic sources, the typical user-provided parameter is a combination of the PCR, SCR, and the maximum burstiness over the entire duration of the traffic. Second, the difficulty in accurately defining PCR arises from the requirement that the smallest time interval must be specified over which the PCR is computed which, in the limit, will approach zero or the network's resolution of time. Third, the literature does not contain any reference to a scientific principle underlying these guarantees. Under these circumstances, the issue of providing QoS guarantees in the real world, through traffic controls applied on a per call basis, is rendered uncertain. This paper adopts a radically different, high level approach to the issue of QoS guarantees. It aims at uncovering through systematic experimentation a relationship, if any exists, between the key high level user traffic characteristics and the resulting QoS measures in a realistic operational environment. It may be observed that while each user is solely interested in the QoS of his/her own traffic, the network provider cares for two factors: (1) Maximize the link utilization in the network since links constitute a significant investment, and (2) ensure the QoS guarantees for every user traffic, thereby maintaining customer satisfaction. Based on the observations, this paper proposes a two-phase strategy. Under the first phase, the average "link utilization" computed over all the links in a network is maintained within a range, specified by the underlying network provider, through high level call admission control, i.e. by limiting the volume of the incident traffic on the network, at any time. The second phase is based on the hypothesis that the number of traffic sources, their nature--audio, video, or data, and the bandwidth distribution of the source traffic, admitted subject to a specific chosen value of "link utilization" in the network, will exert a unique influence on the cumulative delay distribution at the buffers of the representative nodes and, hence, on the QoS guarantees of each call. The underlying thinking is as follows. The cumulative buffer delay distribution, at any given node and at any time instant, will clearly reflect the cumulative effect of the traffic distributions of the multiple connections that are currently active on the input links. Any bounds imposed on the cumulative buffer delay distribution at the nodes of the network will also dominate the QoS bounds of each of the constituent user traffic. Thus, for each individual traffic source, the buffer delay distributions at the nodes of the network, obtained for different traffic distributions, may serve as its QoS measure. If the hypothesis is proven true, in essence, the number of traffic sources and their bandwidth distribution will serve asa practically realizable high level traffic control in providing realistic QoS guarantees for every call. To verify the correctness of the hypothesis, an experiment is designed that consists of a representative ATM network, traffic sources that are characterized through representative and realistic user-provided parameters, and a given set of input traffic volumes appropriate for a network provider approved link utilization measure. The key source traffic parameters include the number of sources that are incident on the network and the constituent links at any given time, the bandwidth requirement of the sources, and their nature. For each call, the constituent cells are generated stochastically, utilizing the typical user-provided parameter as an estimate of the bandwidth requirement. Extensive simulations reveal that, for a given link utilization level held uniform throughout the network, while the QoS metrics--end-to-end cell delay, jitter, and loss, are superior in the presence of many calls each with low bandwidth requirement, they are significantly worse when the network carries fewer calls of very high bandwidths. The findings demonstrate the feasibility of guaranteeing QoS for each and every call through high level traffic controls. As for practicality, call durations are relatively long, ranging from ms to even minutes, thereby enabling network management to exercise realistic controls over them, even in a geographically widely dispersed ATM network. In contrast, current traffic controls that act on ATM cells at the UNI face formidable challenge from high bandwidth traffic where cell lifetimes may be extremely short, in the range of µs. The findings also underscore two additional important contributions of this paper. First, the network provider may collect data on the high level user traffic characteristics, compute the corresponding average link utilization in the network, and measure the cumulative buffer delay distributions at the nodes, in an operational network. The provider may then determine, based on all relevant criteria, a range of input and system parameters over which the network may be permitted to operate, the intersection of all of which may yield a realistic network operating point (NOP). During subsequent operation of the network, the network provider may guide and maintain the network at a desired NOP by exercising control over the input and system parameters including link utilization, call admittance based on the requested bandwidth, etc. Second, the finding constitutes a vulnerability of ATM networks which a perpetrator may exploit to launch a performance attack.

High performance scientific and engineering applications running on clusters have different communication requirements. Current cluster configurations typically provide multiple network interfaces per node and multiple interconnections among nodes. However, transport protocols such as TCP do not utilize existing multiple network interfaces to enhance communication performance. This paper introduces a new configurable communication model utilizing multiple interconnections. The model adds mechanisms to manage and enhance the overall communication performance of clusters. These configurations include the use of parallel message transfers, the separation of the transfer channels between small messages and large messages, and load balancing among the channels. The main advantages of the model are: (1) providing a flexible, enhanced network infrastructure, (2) hiding the technical details of the heterogeneous network resources from the applications, and (3) providing an easy and flexible way to extend the network capacities for specific nodes. To illustrate the advantages and performance enhancements of the model, a prototype was implemented to experimentally evaluate the cluster network performance, which showed considerable gains.

The OptIPuter is a radical distributed visualization, teleimmersion, data mining and computing architecture. Observing that the exponential growth rates in bandwidth and storage are now much higher than Moore's Law, this major new project of several universities--currently six in the US and one in Amsterdam--exploits a new world of computing in which the central architectural element is optical networking. This transition is caused by the use of parallelism, as in supercomputing a decade ago. However, this time the parallelism is in multiple wavelengths of light, or lambdas, on single optical fibers, creating a LambdaGrid. Providing applications-centric middleware to control the LambdaGrid on a regional and global scale is a key goal of the OptIPuter and StarLight Optical Switching projects.

This paper presents an architecture for content internetworking, which we call CRN (Content Routing Network) architecture. The CRN architecture is different from other content internetworking architectures in many respects: the peering of authentication, authorization and accounting systems, hierarchical and policy-driven request routing, and the web-based system to interconnect distinct CDNs. Both requirements and functional architecture of CRN are presented, followed by the description of its prototypical implementation. CRN is designed to satisfy both content provider's service requirements and service provider's economic/operational requirements. A prototypical implementation has been deployed successfully under one of the biggest live-streaming experiments.

The modern network service of finding the optimal path subject to multiple constraints on performance metrics such as delay, jitter, loss probability, etc. gives rise to the multi-constrained optimal-path (MCOP) QoS routing problem, which is NP-complete. In this paper, this problem is solved through both exact and heuristic algorithms. We propose an exact algorithm E_MCOP, which first constructs an aggregate weight and then uses a K-shortest-path algorithm to find the optimal solution. By means of E_MCOP, the performance of the heuristic algorithm H_MCOP proposed by Korkmaz et al. in a recent work is evaluated. H_MCOP only runs Dijkstra's algorithm (with slight modifications) twice, but it can find feasible paths with a success ratio very close to that of the exact algorithm. However, we notice that in certain cases its feasible solution has an unsatisfactorily high average cost deviation from the corresponding optimal solution. For this reason, we propose some modified algorithms based on H_MCOP that can significantly improve the performance by running Dijkstra's algorithm a few more times. The performance of the exact algorithm and heuristics is investigated through computer simulations on networks of various sizes.

The development of efficient quality of service (QoS) routing algorithms in a high-speed network environment is a very important and at the same time very difficult task due to the need to provide divergent services with multiple QoS requirements. Recently heuristic algorithms based on Lagrange relaxation techniques have been proposed to resolve the contradiction between the time complexity and the quality of solution. In this paper, we investigate the performance of two heuristic algorithms, LR_DCLC and NR_DCLC, for the delay-constrained least-cost (DCLC) routing problem. Algorithm LR_DCLC is based on linear relaxation, while algorithm NR_DCLC, which is proposed in this paper, is based on nonlinear relaxation. A large number of simulations demonstrate that even though both algorithms have very good performance, NR_DCLC can obtain much better solutions than LR_DCLC by running Dijkstra's algorithm on average a few more times, especially in the case when the optimal solutions are hard to find.

For future adaptive networking services, it will be effective if autonomous agents utilize widely spread information that continuously changes. A communication network should provide a framework for agents to utilize information in a large-scale network and should positively support communications among agents. This paper proposes EcAgent (Environment for Communities of Agents). EcAgent provides agents with access to communities that are loose collections of agents. It gathers global information about communications among all agents and makes the agents share it in communities. It makes possible adaptive networking services that use various types of information. We implemented the EcAgent execution environment and some experimental applications to make an adaptive networking service. We also implemented a simulation system to evaluate how accurately the library recognized the embedded relationships as communities. The results of experiments show that the proposed methods for community recognition work efficiently.

The problem of finding a path with two additive constraints, in particular finding a path that satisfies both the cost and the delay constraints, is called multi-constrained path (MCP) problem in the literature. In this paper, we explore the MCP problem based on the idea of single mixed weight --a mixed weight for each link is first obtained by combining its delay and cost, and then Dijkstra's algorithm is used to find the corresponding shortest path. Given two infeasible paths, it can be theoretically proved that a better path can possibly be found if we choose an appropriate parameter to construct the mixed weight. An approximate algorithm is thus proposed to solve the MCP problem. Theoretical analysis demonstrates that this algorithm can make a correct judgment whether there is a feasible path or not with a very high probability even in the strict case where the delay bound is between the delays of the least delay path and the least cost path, while the cost bound is between the costs of the two paths. On the other hand, the time complexity of this algorithm is very small since it only needs to execute Dijkstra's algorithm a limited number of times. The excellent performance of the proposed algorithm is verified by a large number of experiments on networks of different sizes.

This paper describes the concept of an adaptive network, that is, a network environment that can rapidly and autonomously adapt its behavior according to network conditions and traffic status. The user interface of the adaptive network can access any resource in the network as a memory-mapped I/O device, as if it were attached to the local bus of the user's PC. This network concept has several benefits. From the application development viewpoint, no network related programming is required, and applications do not have to be modified even if the network topologies and protocols are changed. Network maintenance and upgrading can be done anytime without having to worry about the application users, because the network itself is concealed from the applications. In addition, the reconfigurable hardware technology functions as an autonomous network control through the use of a lower-layer protocol. We developed a testbed that makes heterogeneous resources available to users and used it to demonstrate the feasibility of our concept by implementing and running some applications over it.

This paper proposes a credit-based congestion control scheme for multicast communication which employs application-specific processing at intermediate network nodes. The control scheme was designed not only to take advantage of credit-based flow control for unicast communication, but also to achieve flexibility supported by active network technology. The resultant active multicast congestion control scheme is able to meet the different requirements of various multicast applications in terms of reliability and end-to-end latency. The performance of the proposed control scheme was evaluated using both discrete-event simulations and experiments on a prototype active network implementation. The results show that the proposed scheme performs very well in terms of fairness, responsiveness, and scalability. The implementation experiences also confirmed the feasibility of the scheme in practice.

The heterogeneous autonomous decentralized system technology offers a way to integrate different types of context-related autonomous decentralized (sub) systems into a coherent system. The aim of this research is to model and evaluate the communication capacity among the subsystems connected by communication gateways of a heterogeneous autonomous decentralized system. Failures of subsystems and communication gateways in the system are taken into account. We use graphs to represent the topologies of heterogeneous autonomous decentralized systems and use the residual connectedness reliability (RCR) to characterize the communication capacity among its subsystems connected by its gateways. This model enables us to share research results obtained in residual connectedness reliability study in graph theory. Not to our surprise, we learnt soon that computing RCR of general graphs is NP-hard. But to our surprise, there exist no efficient approximation algorithms that can give a good estimation of RCR for an arbitrary graph when both vertices and edges may fail. We proposed in this paper a simulation scheme that gave us good results for small to large graphs but failed for very large graphs. Then we applied a theoretical bounding approach. We obtained expressions for upper and lower bounds of RCR for arbitrary graphs. Both upper and lower bound expressions can be computed in polynomial time. We applied these expressions to several typical graphs and showed that the differences between the upper and lower bounds tend to zero as the sizes of graphs tend to infinite. The contributions of this research are twofold, we find an efficient way to model and evaluate the communication capacity of heterogeneous autonomous decentralized systems; we contribute an efficient algorithm to estimate RCR in general graph theory.

In order to realize universal service access of mobile users on the global telecommunication environment, we have to consider the scalable networking architecute based on the distributed object computing technologies. In this paper, we apply TINA to the promising networking architecture, and propose TINA-based scalable model for realizing personal mobility service, which especially focuses on making access to telecommunication services from any legacy systems that do not support such the new environment, using a downloadable module in an end terminal, as well as considering load balance of a whole system. Our proposed model also focuses on exchanging personal infomation of mobile users between different administrative domains, by separating the already-specified user profile information objects into two parts, which are the domain-common part and domain-specific part, and propose secure exchange procedure.

In this paper, we propose an architecture for environmental information services that make it possible for applications to adapt in dynamically changing environments. These services provide information necessary for applications to adapt to its environment. Unlike other information services, the information provided includes not only raw information but also abstract or policy-applied information. The variety of information enables applications to choose suitable level of information. A simple adaptive application can use highly abstract information instead of detailed raw information required by complicated adaptive applications. The policy-applied information enables adaptive applications to share decisions by the user and cooperate among them. Applications can efficiently adapt to changes in the environment since our architecture provides notification of these changes. This notification does not disturb applications since the conditions for the notification are controlled by each application. We apply the proposed architecture to a mobile internetworking environment and present a prototype implementation of environmental information services called CM1. We also discuss our primary evaluation of CM1 with the Personal File System, which is a network file system with dynamic adaptation features, for mobile internetworking environments.

The Multiwavelength Optical Networking (MONET) program consists of a consortium of industrial partners, working together with the intent to demonstrate the key capabilities needed for configurable WDM networks. This involves integrating WDM technologies with optical switching technologies to provide a managed, high capacity, national scale WDM server layer to transport optical signals transparently across multiple interworking subnetworks.

This paper describes recent standardization activities on optical networking and relevant issues which are conducted in ITU-T. The organization is first described, then, several recommendations both which are already recommended and expected to be completed in a couple of years.

In this paper, we propose a scalable service networking architecture as a TINA-like environment for providing flexibly various mobility services. The proposed architecture provides an environment that enables the advent of service providers and rapidly introduces multimedia applications, considering networks scalability. For supporting customized mobility services, this architecture adopts a new service component, which we call Omnipresent Personal Environment Manager (OpeMgr). In order to support mobile users who move between heterogeneous networks, for instance, between the TINA-like environment and the Internet environment, we propose a structure of a gateway. In addition, the proposed architecture uses the fixed and mobile agent approaches for supporting the user's mobility, and we evaluated their performances with comparing those approaches.

The progress of multimedia applications for education, research, social welfare and commerce is generating a lot of interest in the potential of a combination of satellite networking and Internet technology. The combination is particularly attractive as a low cost solution in regions which are large and sparsely populated. In 1991, aiming at networking the Pan-Pacific region, the PARTNERS (Pan-Pacific Regional Telecommunications Network Experiment and Research by Satellite) project was initiated. In this project, the major target was to construct a satellite-based network infrastructure to support education, research and so on in the Pan-Pacific region. As a part of PARTNERS the MEISEI-NET (Multimedia EducatIon System using satellite ETS-V and InterNET) project was started to evaluate the utility of satellite networking for education and reserch and, to investigate the feasibility of expanding the reach of the Internet using the PARTNERS infrastructure. MEISEI-NET focussed on (1) low start-up cost, (2) open access to the rich information resources on the Internet, (3) use of network to support education and research, and , (4) development and distribution of software for MEISEI-NET users. The construction of MEISEI-NET will be detailed followed by a report on its usage and the effects of this network. To support and manage MESEI-NET operations, we developed and deployed SNMP-based intelligent network management system. It offered fault detection and notification. This made the MEISEI-NET robust and practical despite of the satellite's (ETS-V) drift-problem. Students and researchers of universities from different countries participated in and benefited from MEISEI-NET until March 1996.