Michael Post

A Distributed Evolutionary Algorithm for Reorganizing Network Communications

The Distributed Evolutionary Algorithm (DEA) presented in this paper produces efficient TDMA schedules for communication in an environment of changing network topology by passing topological and traffic information among nodes communicating on the schedule. As information is gained in this way more nodes are added to the communicating group until the entire network is communicating on a common schedule. The key feature of the DEA presented is that the reorganization phase is fast, requiring a number of TDMA slots only on the order of the number of nodes. A secondary feature is that some network communication can take place, and thus some of the traffic requirements can be satisfied, even while the network is being reorganized. Numerical results are presented for a number of randomly generated networks which show how the algorithm performs.

Towards hitless reconfiguration in WDM optical networks for ATM transport

This paper proposes a methodology for reconfiguring a wavelength division multiplexed (WDM) optical network to adapt to changing traffic requirements at the ATM layer. The ATM network layer is reconfigured to adapt to changes in traffic patterns. These changing ATM network topologies or connectivities are known and are used to design the network. The proposed method sizes the ATM switches and then assigns wavelengths between pairs of ports so as to support the required ATM network topologies and their reconfiguration in a hitless manner.

Connection management for multiwavelength optical networking

This paper describes the connection management research done by the Network Control and Management (NC&M) task force under the multiwavelength optical networking (MONET) program. MONET is sponsored by the Defense Advanced Research Project Agency (DARPA) the US Government Department of Defense, with participation from Bellcore, AT&T, Lucent Technologies, several government agencies, and regional Bell operating companies. MONETs vision is to develop a flexible reliable high-capacity high-performance cost-effective national optical network based on multiwavelength fiber optic technology. As an important component in realizing this vision, the MONET program includes the architecture and design of a prototype network control and management system for MONETs reconfigurable wavelength-division multiplexing (WDM) all-optical network. The primary objectives of the prototype research work are to develop the architecture and framework for managing national-scale transparent reconfigurable WDM optical networks and to demonstrate the feasibility of the NC&M prototype system in a field experiment network in Washington, DC. This prototype system allows the program participants to conduct experiments and gain experience in the management and operations of reconfigurable optical networks. This paper describes the connection management aspects of the prototype system, addressing issues such as the management architecture, information model, and provisioning algorithms of the prototype management system.

The manager/agency paradigm for distributed network management

The manager/agent model has been the de facto paradigm used for designing network management systems. However, as the network size grows and new network and computing technology emerges, consistency, flexibility, reliability, and scalability issues of distributed network management systems might not be fully addressed within the model. This paper proposes an extension to the manager/agent model, called the manager/agency paradigm, to address these issues. In the manager/agency paradigm, a manager performs management functions through an agency of cooperating management entities, (lower-level) managers and/of agents, which are collaboratively working together to achieve consistent management goals. Agencies may recursively constitute other agencies, based on the layering and partitioning principles, to scale with ever growing network size. They may be organized flexibly according to different management functionalities, goals and policies, and may present a single management view to their (upper-level) managers by hiding management complexity. Inside an agency, various fault tolerant schemes, such as active replication or primary-backup approach, may be used to implement reliable management functionality transparently to its manager. The paper begins by describing the anatomy and semantics of an agency and comparing the manager/agency model against the manager/agent model. It then demonstrates its applications to address the consistency and reliability issues of network management problems. Finally, applications of the manager/agency model to the broader, distributed enterprise management is discussed.

Scheduling multihop CDMA networks in the presence of secondary conflicts

This paper presents an algorithm for producing near-optimal conflict-free schedules for networks operating under code division multiple access (CDMA). A procedure for finding a lower bound on the length of such schedules is also presented. The presence of both primary and secondary conflicts (due to imperfectly orthogonal CDMA codes) are accounted for by these algorithms. The complexity of both algorithms is analyzed and computational experience with both procedures is presented. Using the lower bound, it is shown that the heuristic is effective. The complexity analysis demonstrates that it is efficient enough to use in networks of realistic size, even when the schedules must be produced in real time.

Distributed scheduling of CDMA networks with minimal information

An algorithm which produces conflict-free communication schedules in mobility multihop radio networks is presented. These schedules are produced in a completely distributed manner. The algorithm is based on a globally known permutation on the nodes of the network. As a result the only knowledge needed on the part of individual nodes is the number of nodes in the network. This permutation guarantees that conflict-free schedules can be produced in a distributed manner. Two extensions to the basic permutation are discussed. The first enables neighboring nodes to enhance their communication schedules in a fast, robust, distributed manner. The second extension allows the algorithm to operate in the presence of secondary conflicts. >

Network control and management of reconfigurable WDM all-optical network

To meet the demand for huge bandwidth created by the multimedia communications revolution, there is an urgent need to dramatically upgrade the existing telecommunications infrastructure. In todays infrastructure, fiber-optics is used in point-to-point transmission links, with all networking and services functions done electronically. To upgrade such an infrastructure in a cost-effective way, Wavelength Division Multiplexing (WDM) technology may be used to transmit multiple optical channels per fiber. This way, network capacity may grow in response to increased demand. In addition, with the reconfigurability of WDM networks, network configuration may adapt in response to changing traffic patterns. To ensure successful deployment and operations of WDM networks, it becomes essential to have a comprehensive network control and management system which is versatile, robust, and scaleable. In this paper, we describe a network control and management system prototype developed under the Multiwavelength Optical Networking (MONET) project funded by the US governments Defense Advanced Research Project Agency. The prototype system provides configuration, connection, fault, and performance management functionalities for reconfigurable WDM all-optical networks. The design of the prototype system follows the TMN logical layered architecture, and it is implemented on a fault-tolerant, CORBA-compliant distributed object computing platform which facilitates robust and scaleable operations.

A framework for adding real-time distributed software fault detection and isolation to SNMP-based systems management

We consider the problem of fault detection and isolation in systems that consist of real-time distributed cooperating processes. A framework for adding fault detection and isolation capabilities to SNMP-based distributed management systems is presented. The framework revolves around the use of a formal specification model of the cooperating processes which we refer to as “local directed graphs”. We describe the local directed graph model, and a fault monitoring and isolation architecture that implements the framework. In doing so, we address the problem of the size of our formal description and also show that this architecture is suited to the management of internetworks. Lastly, we present an example illustrating the operation of the architecture.

Network management and control of protocols

The authors describe some of the special requirements placed on the protocols that must function in a multidomain environment. Taking these requirements into consideration, they describe a novel protocol structure called a bipartite protocol, which is based on the use of tokens. They show that the use of bipartite protocols makes it possible to perform a number of functions, namely to systematically explore subgraphs of the global state diagram representing the protocol, with predetermined numbers of messages in the channel queues, thereby facilitating the analysis of protocols; to generate tightly synchronized protocols or sections of protocols, thereby controlling the build up of queues in the channels; and to implement sections of network control systems using the tokens in the bipartite protocol structure. Examples are given for X.75, illustrating both the analysis of protocols and the control of channel contents.<<ETX>>

Fault Isolation in Communication Networks with Local Directed Graphs

This paper presents an approach to real-time fault detection and isolation in communication networks due to erroneously implemented software. Unlike other approaches to real-time fault detection and isolation, the approach using local directed graphs presented in this paper does not require a monitoring system for every element of the communications network.