Using the Concept of Intelligent Agents in Fault Management of Distributed Services

This paper proposes the application of conceptsfrom the area of intelligent agents to overcomedeficiencies of existing management architecturesregarding distribution of functionality and flexibility. Its main contribution is the proposal of amethodology for a flexible, distributed realization ofcomplex management tasks. The main application areas aredistributed services which are complex pieces of software, distributed across variousheterogeneous end systems in a network. Mostly, theyrely on the provision of other services as well. Theapproach relies on well-known concepts, such ascooperative distributed problem solving and intelligentagents, and offers a framework to combine these twoconcepts, providing a step on the roadmap to a flexible,distributed management architecture. The assessment of the approach is displayed throughout thepaper by scenarios from the area of nontime criticalfault management.     

A Framework for Designing Distributed Management Applications

This paper presents a methodology for thespecification and design of multi-domain managementsystems that control, monitor and administer users andservices in a distributed telecommunication environment. It presents a framework, which serves as ablueprint for the design and implementation ofdistributed management applications. It defines a set ofconcepts and methods for guiding the analysis, design, and implementation phases of managementapplications, which are distributed over heterogeneoussystems. The framework is based on concepts ofTelecommunications Management Networks and OpenDistributed Processing, and is partially developed withinthe Pan-European Reference Configuration for IBC(Integrated Broadband Communication) Services Management(PRISM) project.     

Performance and Efficiency in Distributed Enterprise Management

This paper is motivated by the increasing needfor scaleable, distributed management architectures forintegrated network, system and application managementwithin the enterprise network environment. Such integration, extension and wide area deploymentof management functionality impose heavy performancerequirements and produce increased management traffic.Aiming to minimize this traffic and the overall response time, we propose a distributedhierarchical caching scheme that attempts to takeadvantage of the diverse consistency requirements ofmanagement applications. We define coherency conditionsand update policies, identify the appropriateinteraction semantics, and discuss an SNMP-basedimplementation. In order to evaluate the proposed modeland to quantify the expected performance gains weconstruct a simple queuing model that provides analyticalresults on the improvement of response time and thereduction of management traffic. Finally, the analysisof experimental results provides some insight on performance improvement for specific classes ofmanaged objects.     

TINA and the Software Infrastructure of the Telecom Network of the Future

The overall goal of the TelecommunicationsInformation Networking Architecture (TINA) is to defineand specify a software architecture for thetelecommunications infrastructure which encompassesnetworking, network services, and operations, to handle theincreasingly distributed nature of software in networks.The purpose of this paper is to describe how TINA may bepositioned in the software infrastructure of future telecommunications networks, with thefocus on the infrastructures of Public TelecommunicationOperators (PTOs). An architecture for a new PTO networkinfrastructure is proposed and the needs that this architecture satisfies are discussed.The concept of the programmable network isintroduced to satisfy the PTO need to quickly trial anddeploy new services. Five aspects of this targetarchitecture are discussed, namely: a client/serverarchitecture, a service creation architecture, anapplication architecture, a PTO federation architecture,and a new management architecture. The potential role of TINA in each aspect is described, and thepotential benefits of TINA to PTOs aresummarized.     

A Re-Engineering, Methodology for Cooperative Management of Enterprise Networks

With the increasing implementation by networkedmission-critical applications, an enterprise network isbecoming the lifeline of an organization. Massiveinvestments are being made in the modernization of enterprise networks of diverse serviceorganizations, such as telecommunications andhealth-care providers. Organizations need to adapt tochanging business environment, such as the deregulationof telecommunication services. This necessitates a seamlessintegration of business management processes withenterprise network management processes. Hence there isa need for the formulation of new methodologies for there-engineering of management solutions (with focus onintegrated business processes), though most present daysolutions concentrate on the management of networkequipment only. This paper presents a new methodologyfor the reengineering of the management processesfor enterprise networks, based on Computer SupportedCooperative Work (CSCW) techniques.     

An Open Secure Mobile Agent Framework for Systems Management

The Mobile Agent (MA) technology is gainingimportance in the distributed management of networks andservices for heterogeneous environments. MA-basedmanagement systems could represent an interestingalternative to traditional tools built upon theclient/server model, either SNMP- or CMIP-based. Twomain requirements currently limit the acceptance of MAsolutions for management: the need of interoperabilityand the request for security. Without security,management systems cannot suit global untrustedenvironments, such as the Internet; withoutinteroperability, they cannot interact with existingtools and legacy systems. The paper describes an MA-basedmanagement system with security and interoperability asthe two main design objectives. It is an open managementframework that grants interoperability by providing compliance with CORBA, the most diffusedstandard in the area of object-oriented components. Inaddition, it is based on a thorough security model andprovides a wide range of tools and mechanisms to build and enforce flexible securitypolicies.     

An ATM-Based Connection Management Architecture for Telecommunications Services

New open service architectures are now emerging.An important part of these architectures is theconnection management function. It enables theestablishment, control and release of semipermanentconnections requested by telecommunications services suchas virtual private networks. This paper presents thespecification, design and implementation of thisconnection management function in our OAMS Open Service Architecture. The proposed architecture isdeployed over a distributed processingenvironment.     

Temporal Network Management Information Model and Services

The paper addresses the issue of time as anattribute of the network management information. Thetemporal dimension is incorporated in the managementinformation model proposed by the Internet Engineering Task Force (IETF), as it is described by theStructure of Management Information (SMI). The core ofthe proposed network management information model is theTemporal Management Information Base (TMIB), a conceptual representation of the diachronicbehavior of network resources. The temporal networkmodel also includes the definition of a specificarchitecture and a number of temporal services. Theexistence of such services facilitates the development oftime-reference applications in different functionalareas.     

The Role Multimedia of Multicasting in Managing Interactive Distance Learning Systems

This paper discusses the important role ofmulticasting in designing, implementing, and managinginteractive multimedia distance learning systems. Thisis achieved in the context of IRI, an Interactive Remote Instruction system for distance learningbuilt at Old Dominion University. IRI is anInternet-based system which integrates continuousmultimedia, shared applications and a variety ofmulti-user collaborative utilities. In this paper, weconcentrate on the process architecture and dynamicmulticast group handling as they pertain to managingmultimedia resources, and show how they supportrobustness and short response time to user actions. IRIuses raw IP multicasting for audio and video streams andreliable multicasting for resource management and datasharing. The system is scalable (uses multicast for inter-process communication) andexpandable(partitioned into a setofautonomousbutcooperating components).     

A Framework for Predictive Network Management of Predictive Mobile Networks

There is a trend toward the use of predictivesystems in communications networks. At the systems andnetwork management level predictive capabilities arefocused on anticipating network faults and performance degradation. Simultaneously, mobilecommunication networks are being developed withpredictive location and tracking mechanisms. Theinteractions and synergies between these systems presenta new set of problems. A new predictive network managementframework is developed and examined. The interactionbetween a predictive mobile network and the proposednetwork management system is discussed. The Rapidly Deployable Radio Network (RDRN) is used as aspecific example to illustrate theseinteractions.     

A Model ATM Network and APIs for Service Management on an ATM Network

The architectural reference point for our workis a reference model which associates functionality withoperational planes along three orthogonalabstractedmodels. Within the context of this reference model a service management model is proposed. Themodel makes use of APIs for communicating managementinformation between a management application and serviceentities distributed in an ATM network. The APIs are based upon native ATM APIs that are similarto those that have recently been adopted by the ATMForum. The viability of the model is demonstrated bybuilding ATM multimedia services with service management capabilities. The model and the APIs have beenapplied to services executing in a national ATM network.Although the work is motivated by an ATM network, theservice management model is general, and may be applied to other networks as well. Some openissues in service management are brought out.     

Connectivity Infrastructure and Components for POCT Environments - Overall Infrastructure

Connectivity infrastructure, hardware and software components for point-of-care testing (POCT) environments at medium-to-large sized hospitals with expected number of POCT instruments in the order of hundreds or thousands are described. The instruments include both network-ready and non-network-ready devices. The latter are connected to the network by means of a hardware-based Instrument Network Adapter. Instrument messages are converted to standardized form and, depending on message content, are routed and delivered to appropriate destinations by a software-based Message Routing System consisting of a Message Router and a Delivery Agent. Target information systems consist of departmental and central information systems.     

Management Issues of a Mobile Agent-Based Service Environment

This paper introduces the concepts of a servicemanagement system for mobile agent-based services. Incontrast to several other approaches in whichagent-based applications are used to handletraditional network or service managementaspects, this paper describes a way to manage thedistributed agent environment itself. The proposedapproach was designed and developed by using OSI SystemsManagement concepts as a starting point, which are modified andadjusted in order to fulfill the new requirementsassociated with the mobile agent paradigm. However, thispaper does not cover integration or co-existence of the introduced agent management system (AMS)and traditional OSI management systems. Instead, theentire AMS environment is based on mobile agenttechnology. At present, this approach is validated inthe context of the European research project MobileAgent Environments in Intelligent Networks(MARINE).     

Contrasting OSI Systems Management to SNMP and TMN

OSI systems management is a set of standards forthe management of open systems. Its acceptance by theindustry has been slower than that of SNMP, due to itsrelative complexity. However, it is gathering more and more momentum due to the increasingavailability of development tool kits and also due tothe real world implementations which have followed theuse in TMN. This paper contrasts OSI systems management to SNMP and TMN. It identifies the salientcharacteristics of OSI systems management and uses themas bases for the comparison. Aspects of X.25 managementare used for illustration purpose. OSI system management is half way between SNMP and TNM. It offersmore than SNMP, but less than TMN. SNMP, OSI systemsmanagement and TMN tackle different classes ofmanagement problems. None is a panacea andinteroperability is the key issue.     

A Model-Based Performance Management Tool for ATM and Frame Relay Networks

In this paper we describe a network modelingapproach intended to assist in the performancemanagement, design, and optimization of broadbandtraffic networks. Switch and source models, as well asrouting optimization and decision support algorithmshave been integrated in a prototype software tool,called DATANMOT (Data Network Modeling and OptimizationTool). The switch models developed are based on standard Frame Relay and ATM switch implementations.Specifically, an analytical model of the Fujitsu FETEX150 ATM switch is described here in detail. Fluid-flowapproximation methods were used for performance evaluation, with computational complexity lowenough for near-real time applications. As a result,given the network configuration and input traffic, anevaluation of the quality of service can be derived and used in optimal routing, admission controland network planning. These techniques have beenincorporated in our modeling tool to demonstrate themodel-based approach to network management. In addition, all configuration, modeling, and managementfunctions of the software tool are supported by agraphical user interface, and a databasesystem.     

A Generic Architecture Multimedia Service and Management

A multimedia service architecture is constructedby using a generic component model. In order to derivespecific components, a generic service agent and genericservice component architectures are developed. Using generic agent and the service components,a TINA-C conformant object-oriented hierarchical genericservice component is developed and presented. As anexample, a multimedia QoS service and management architecture is presented.     

一个基于CORBA的分布式网管系统模型ACORBA-BasedModelfor

新一代的分布式网管系统应具有良好的可扩展性、可集成性等特征。本文分析了当前主流网管系统的不足之后,给出了基于ＣＯＲＢＡ的分布式网管系统模型,并进一步阐述了分布对象技术（ＣＯＲＢＡ）在网管领域的应用优劣。     

A Survey of Developments in Distributed Data Base Management Systems

Distributed data base management systems extend the capacities of computing systems to give many people access to a wide range of information. These systems, which arise from the combination of recent important advances in computer networks with general-purpose data base management systems, should shortly represent another significant step in commercial system development. So it is not surprising that they have been the focus of much research and development activity in both the academic and industrial environments.     

分布的共享主存多机系统综述

分布的共享主存克服了紧耦合共享主存多处理机及分布式多机系统的缺点,既有共享的主存,又具有可扩展性。本文总结了设计分布的共享主存所涉及到的技术选择,分析了实现上的关键问题,并对具有代表性的系统进行了分析,最后总结了现有分布的共享主存的特点,提出了新的研究方向。     

The Second Generation of the Laboratory Information System and the Laboratory Automation System in Osaka City University Hospital

The first total clinical laboratory system (TCLA) in the Osaka City University Hospital was introduced in 1993. After six years of operation, it was recently replaced by a new system. The goals of this replacement were as follows: 1. Improve the analytical performance. 2. Make the system operate more efficiently. 3. Improve the quality of laboratory analyses. We successfully reduced the labor required to operate the laboratory; made the laboratory reports quicker; reduced the number of retraction and revalidation of the results; and minimized the system downtime.