The demand assigned capacity management (DACM) problem in IP over optical (IPO) network aims at devising efficient bandwidth replenishment schedules from the optical domain conditioned upon traffic evolution processes in the IP domain. A replenishment schedule specifies the location, sizing, and sequencing of link capacity expansions to support the growth of Internet traffic demand in the IP network subject to economic considerations. We develop mathematical models to address the DACM problem in IPO networks within a deterministic context based on a class of inventory management replenishment methods. We apply the technique to IPO networks that implement noncapacity adaptive routing in the IP domain and analyze the performance characteristics in terms of minimizing cumulative replenishment cost over an interval of time. This study represents an application of integrated traffic engineering which concerns collaborative decision making targeted towards network performance improvement that takes into consideration traffic demands, control capabilities, and network assets at different levels in the network hierarchy.

Provides an overview of the technical articles and features presented in this issue.

Internet traffic engineering is concerned with the performance optimization of IP networks. Multiprotocol Label Switching (MPLS) is an important technology that supports the Internet traffic engineering function. This paper considers a network performance optimization problem related to traffic engineering over MPLS. We investigate the use of a dynamic traffic partitioning and assignment methodology to adaptively map ingress traffic into several parallel Label Switched Paths (LSPs) in MPLS based IP networks. We present a stochastic framework for the traffic partitioning problem. Within this framework, a set of parallel edge disjoint LSPs is modeled by parallel queues and a partitioning algorithm is devised for different service classes that is adaptive to the prevailing state of the network. The performance of this approach is illustrated with numerical examples.

This paper describes a framework that addresses the important problem of optimal location of peer points for data exchange between autonomous systems in the Internet. The decision objective is to minimize the cost of peering, and at the same time improve the efficiency of inter-domain traffic flows. A combinatorial optimization formulation is used to model this problem, and location theory is applied to construct an integer zero-one mathematical program. Solution methodologies for the resulting mathematical program are surveyed and practical implementation considerations are discussed.

RSVP (resource reservation protocol) is a receiver oriented resource reservation setup protocol targeted for integrated services packet networks (ISPNs). RSVP has a number of desirable attributes which make it a leading candidate for Internet standardization. These attributes include flexibility, robustness, and scalability. However, RSVP does not adequately address the issue of resource reservation for mobile hosts. This paper highlights the limitations of RSVP in mobile environments and proposes extensions to obviate these deficiencies. The basic characteristics of the proposed augmentations include: (1) quiescent resource reservations, (2) virtual receivers, (3) intelligent pre-allocation of resources based on analysis of user mobility patterns, and (4) predictive look-ahead dynamic dormant multicast trees.

Future distributed systems are expected to support mo-bile computing through wireless extensions. In such environments, the spatial location of system resources, such as servers, can change dynamically as a result of host mobility. Consequently, a search or discovery cost is incurred when-ever a client requests use of a mobile resource whose location is unknown. In addition, because of the acute dependence of radio link performance on the characteristics of indoor operating environments, the Quality of Service experienced by a client that utilizes a mobile resource can depend on the resource’s specific location. The work reported in this paper proposes and evaluates a statistical framework for resource discovery in distributed computing environments with mo-bile hosts. Specifically, taking the peculiar attributes of wireless segments of the distributed system into account, we investigate the interplay between the cost of search for po-tentially mobile resources and the utility or benefit accrued to clients that require such resources. This interplay has im-portant ramifications for resource allocation in distributed mobile computing environments.