Abstract: This manuscript provides the first systematic test of a formal theory of managing government programs in a network context (see O’Toole and Meier 1999) Using data from several hundred school districts in Texas, we create a measure of network management that reflects the time school superintendents interact with several sets of significant actors in the environment We find that network management is not only related to overall organizationalit has positive impacts even in the presence of a lagged dependent variable Further, management appears to interact with other organizational resources in a nonlinear manner to further augment organizational performance Although public school networks are not as complicated or as populated as those of many other public organizations, the findings suggest how management might matter in these other networks

Abstract: Accurate short-term traffic flow forecasting has become a crucial step in the overall goal of better road network management. Previous research [H. Kirby, M. Dougherty, S. Watson, Should we use neural networks or statistical models for short term motorway traffic forecasting, International Journal of Forecasting 13 (1997) 43–50.] has demonstrated that a straightforward application of neural networks can be used to forecast traffic flows along a motorway link. The objective of this paper is to report on the application and performance of an alternative neural computing algorithm which involves ‘sequential or dynamic learning’ of the traffic flow process. Our initial work [H. Chen, S. Clark, M.S. Dougherty, S.M. Grant-Muller, Investigation of network performance prediction, Report on Dynamic Neural Network and Performance Indicator development, Institute for Transport Studies, University of Leeds Technical Note 418, 1998 (unpublished)] was based on simulated data (generated using a Hermite polynomial with random noise) that had a profile similar to that of traffic flows in real data. This indicated the potential suitability of dynamic neural networks with traffic flow data. Using the Kalman filter type network an initial application with M25 motorway flow data suggested that a percentage absolute error (PAE) of approximately 9.5% could be achieved for a network with five hidden units (compared with 11% for the static neural network model). Three different neural networks were trained with all the data (containing an unknown number of incidents) and secondly using data wholly obtained around incidents. Results showed that from the three different models, the ‘simple dynamic model’ with the first five units fixed (and subsequent hidden units distributed amongst these) had the best forecasting performance. Comparisons were also made of the networks’ performance on data obtained around incidents. More detailed analysis of how the performance of the three networks changed through a single day (including an incident) showed that the simple dynamic model again outperformed the other two networks in all time periods. The use of ‘piecewise’ models (i.e. where a different model is selected according to traffic flow conditions) for data obtained around incidents highlighted good performance again by the simple dynamic network. This outperformed the standard Kalman filter neural network for a medium-sized network and is our overall recommendation for any future application.

Abstract: Accurate network topology information is important for both network management and application performance prediction. Most topology discovery research has focused on wide-area networks and examined topology only at the IP router level, ignoring the need for LAN topology information. Recent work has demonstrated that bridged Ethernet topology can be determined using standard SNMP MIBs; however, these algorithms require each bridge to learn about all other bridges in the network. Our approach to Ethernet topology discovery can determine the connection between a pair of the bridges that share forwarding entries for only three hosts. This minimal knowledge requirement significantly expands the size of the network that can be discovered. We have implemented the new algorithm, and it has accurately determined the topology of several different networks using a variety of hardware and network configurations. Our implementation requires access to only one endpoint to perform the queries needed for topology discovery.

Abstract: Managing a plurality of network devices on a network, including detecting an address assignment message sent from an address server over the network to a network device, the address assignment message containing an assigned address corresponding to the network device, sending an information request message, in response to the detection of the address assignment message, over the network from the network management device to the network device, the information request message containing the assigned address corresponding to the network device, receiving information, in response to the information request message, from the network device, and creating an entry corresponding to the network device in a device management directory, the entry containing the assigned address corresponding to the network device and the information received from the network device.

Abstract: Multirate multicasting, where the receivers of a multicast group can receive service at different rates, is an efficient mode of data delivery for many real-time applications. We address the problem of achieving rates that maximize the total receiver utility for multirate multicast sessions. This problem not only takes into account the heterogeneity in user requirements, but also provides a unified framework for diverse fairness objectives. We propose two algorithms and prove that they converge to the optimal rates for this problem. The algorithms are distributed and scalable, and do not require the network to know the receiver utilities. We discuss how these algorithms can be implemented in a real network, and also demonstrate their convergence through simulation experiments.

Abstract: The illustrative embodiment of the present invention provides a method for automating the allocation of network storage. The method of the present invention enables a system administrator or other authorized user to set policies for network storage. Available storage locations are dynamically located by attribute. The network storage policy is interpreted and applied to the available storage locations. By automating the storage allocation process, the storage policy is consistently applied without user intervention thus allowing a system administrator to devote more time to other network management responsibilities.

Abstract: The present invention provides a method and apparatus for improving management and network availability by providing out-of-band management channels between network/element management system (NMS) clients and servers. High priority client requests and server notifications may be sent over the out-of-band management channels to ensure fast response times. In addition, periodic roll calls between NMS clients and NMS servers may be executed over the out-of-band management channels to allow for quick discovery of any disconnects and reclaiming associated client resources. Further, periodic roll calls may be conducted between the NMS servers and the network devices to which they are connected, and if a server discovers that a network device has gone down, it may send a high priority notification to appropriate NMS clients over the out-of-band management channels to ensure a fast response by the clients. Sending high priority messages over out-of-band management channels maximizes client/server management availability and, hence, network availability.

Abstract: This invention uses Bayesian techniques to prioritize alerts or alert groups generated by intrusion detection systems and other information security devices, such as network analyzers, network monitors, firewalls, antivirus software, authentication services, host and application security services, etc. In a preferred embodiment, alerts are examined for the presence of one or more relevant features, such as the type of an attack, the target of an attack, the outcome of an attack, etc. At least a subset of the features is then provided to a real-time Bayes network, which assigns relevance scores to the received alerts or alert groups. In another embodiment, a network manager (a person) can disagree with the relevance score assigned by the Bayes network, and give an alert or alert group a different relevance score. The Bayes network is then modified so that similar future alerts or alert groups will be assigned a relevance score that more closely matches the score given by the network manager.

Abstract: A network management approach for use in a vehicle to control activation of electronic control units (ECUs) networked together throughout the vehicle. The ECUs are grouped together by function into virtual networks, with each virtual network including those ECUs used in carrying out a particular control task, such as controlling power windows or automatic headlights. The virtual networks are activated using a messaging protocol that specifies which virtual network is being activated. Periodic messages specifying the virtual network are also used to maintain it active. This permits the ECUs to be maintained in a low power quiescent state when the control functions are not needed, while allowing only those needed for a particular control task to be awakened and maintained in an activated state to carry out their associated control task. An ECU can activate one of the virtual networks by transmitting a wake-up signal followed by a message identifying the virtual network. Each of the ECUs receive this message and, if it is a member of the virtual network being initialized, maintains itself in an active state to carry out the control task associated with the virtual network. The other ECUs return to the quiescent state. Using this approach, an ECU is able to activate only the necessary ECUs for a particular control task without having to know which or how many ECUs are involved in performing the task.

Abstract: Apparatus and method for distributed object filtering of objects transferred between a client and server. The objects may be alarms transferred between a network management server and one or more network management applications. The apparatus and method further includes applying a plurality of policy-based filters to the alarms. The filters may be named and stored in a database, and application of the policy-based filters may be scheduled for different times. The same policy-based filters may be applied to one or more multiple network management applications. The system and method provides increased performance and reliability of passing objects from a server to a client application.

Abstract: In survivability management systems, some management entities reside on application hosts that are not necessarily trustworthy. The integrity of these software entities is essential to the security of the network management scheme. In this talk, I present a novel framework to facilitate software security against malicious execution environments. The approach consists of two fundamental techniques: (a) Incorporating diversity in the deployment and the design of the program such that impersonation or intelligent tampering attacks require extensive analysis of the program; and (b) one important aspect of program analysis, namely static analysis, is deterred by the incorporation of aliasing and further degeneration of the program control flow. It is shown that analyzing the transformed programs statically is an NP-hard problem. Theoretic bounds on approximate analysis methods are also provided. The transformations are implemented in a C compiler. Program performance results are presented. Empirical experiments with existing analysis tools showed that static analysis for the transformed programs are hindered to a significant degree.

Abstract: A system and method for managing a network of thin clients is disclosed. The thin clients may be organized into a hierarchy with multiple administrative servers in a hierarchy, each managing one or more thin clients. Updates to thin client configurations may be performed by propagating update information to a top-level master administrative server, which in turn conveys the update information to one or more lower-level remote administrative servers, which in turn convey the update information to their managed thin clients. To simplify network management, the thin clients may be organized into arbitrary clusters, regardless of their position within the hierarchy structure. The hierarchy may also be used to control the propagation of error messages from thin clients. The hierarchy may be implemented using a thin client management program that configures thin clients according to their position within the hierarchy.

Abstract: A hierarchical and distributed control architecture for managing an optical communications network. The architecture includes a line card manager level for managing individual line cards in an optical switch, a node manager level for managing multiple line cards in an optical switch/node, and a network management system level for managing multiple optical switches/nodes in a network. An event manager at the node manager level enables software components that are running at the node manager to register for and receive events, and to post events. The events may be triggered, e.g., by a change in a status of the switch, or an alarm condition. Control architecture functionalities include signaling, routing, protection switching and network management. Furthermore, the network management function includes a topology manager, a performance manager, a connection manager, a fault detection manager, and a configuration manager.

Abstract: A supply chain transaction network. The network includes a customer agent server hosting a collection of intelligent customer agents for interfacing with and extracting information from one or more of a condition monitoring system, a computer maintenance management system and a procurement system and a distributor agent server hosting a collection of intelligent distributor agents for interfacing and communicating with one or more of a supply chain planning system, an inventory management system, a logistic management system, an order management system, an intelligent order fulfillment system, and an equipment knowledge base. An agent network manager and broker routes messages between the distributor agent server and the customer agent server to effect a movement of items within the supply chain.

Abstract: The paper describes work in our project funded by the DARPA Dynamic Coalitions program to design, develop, and demonstrate a system for automatically managing security policies in dynamic networks. Specifically, we aim to reduce human involvement in network management by building a practical network reconfiguration system so that simple security policies stated as positive and negative invariants are upheld as the network changes. The focus of this project is a practical tool to help systems administrators verifiably enforce simple multi-layer network security policies. Our key design considerations are computational cost of policy validation and the power of the enforcement primitives. The central component is a policy engine populated by models of network elements and services that validates policies and computes new configuration settings for network elements when they are violated. We instantiate our policy enforcement tool using a monitoring and instrumentation layer that reports network changes as they occur and implements configuration changes computed by the policy engine.

Abstract: Statistical processing of event outcomes, such as call attempts or handoff attempts, allows reliable generation of performance alarms within a communication network without requiring analysis of historic performance data. Base station controllers might implement such statistical processing so that the controllers themselves rather than other, further removed network management entities generate performance alarms. Attendant advantages include but are not limited to relatively fast and reliable alarm generation using relatively small sample sets. These and other advantages permit detecting and reporting performance alarm conditions more quickly without sacrificing alarm generation reliability.

Abstract: Networking Council Foreword. Acknowledgments. Introduction. A NEW NETWORK MANAGEMENT PARADIGM. New Services, New Requirements. Introduction to Policy-Based Networking. THE COMPONENTS OF POLICY-BASED NETWORKING. What Are Policies? Architectures for Policy-Based Networking. Creating and Managing Policies. The Policy Repository. The Policy Decision Point. Policy Enforcement Points. Monitoring Network Behavior and Policies. An Example of Policy Processing. The Role of Standards in Policy-Based Networking. Directory-Enabled Networks Initiative. APPLICATIONS OF POLICY-BASED NETWORKING. An Introduction to Quality of Service. Policies for Quality of Service. Policies for Network Security. Policy-Based Networking for Enterprises. Policy-Based Networking for Service Providers. Deploying Policy-Based Networking Systems. Appendix A: References. Appendix B: Vendors and Products. Appendix C: Glossary. Index.

Abstract: A client device has access to multiple data communications networks when sending a message to a server. An included network management functionality evaluates on an individual message by message basis a number of factors and selects one of the networks over which the message is to be communicated to the server. The selection process involves having the network management functionality identify a particular selection rule containing a network clause relating to each potentially useable communications network. The particular selection data comprising each network clause are then evaluated in the context of the message transmission to select for the message the particular one of the networks to be used for the communication.

Abstract: A policy based network management (PBNM) system may identify one or more policies associated with a network component (e.g., a network device, a device group, a device subgroup, a user, an application, an end-host, etc.) by identifying one or more policies directly associated with the network component, generating a list of one or more groups to which the network component belongs, and identifying one or more policies associated with each of the groups in the generated list. An aggregated data set (e.g., a hash table or a balanced tree) may be used to store network component identity elements, one or more pointers to a deployed policy tree, and one or more pointers to a network configuration tree. Each identity element in the data set identifies a network component and has an associated network configuration tree pointer and one or more associated deployed policy tree pointers.

Abstract: A network management system provides distributed task and data processing services to management processes by employing distributed autonomous agents. The network management system includes management units (37,38, fig.2) that are associated with network elements (33,34,35,36,fig.2) of the network. The management units (37,38,fig.2) store a dynamically updated model of the corresponding network elements (33,34,35,36,fig.2). The model is used to implement data gathering tasks as well as network control operations. The model includes autonomous agents, each modeling a network element (33,34,35,36,fig.2). The agents communicate with one another and cooperate to facilitate the implementation of a distributed algorithm. The management system provides the management processes with data management as well as one step network control operation services. Accordingly, the management processes do not process a large amount of data or execute step-by-step network control operations.

Abstract: A system and a method for managing information encapsulates the information as objects (34). The objects (34) are related by a degree to other objects (34) in a content network (40). The relationships (42) between the objects (34) may be established and enhanced by various human or automatic means. An attractive user interface (14) facilitates use and management of the network (40) by many users. Access to the content network (40) may be customized for distinct user groups.

Abstract: A security system 63 and an electronic device group 62 to be stored in a home network 6 are controlled by a home server 61 . A home network management facility 5 obtains status information of the electronic device group 62 from home server 61 . Home network management facility 5 displays the status of electronic device group 62 on a display unit of a terminal 1 based on the received status information. Home network management facility 5 also displays a screen for prompting input of a control instruction of the home-located electronic devices that are included in the electronic device group 62 . Based on control instruction received via home server 61 , home network management facility 5 performs remote control of electronic device group 62.

Abstract: There is great interest in exploring SNMP/TCP in addition to the current use of SNMP/UDP due to performance benefits for bulk transfer as well as to simplify management applications. If SNMP is implemented over TCP, then TLS is a natural choice for security. However, it must be demonstrated that the additional overhead associated with TLS is not excessive. We show this by implementing SNMP on a TLS/TCP base and measuring its performance experimentally. The results indicate that the overhead is not excessive; consequently SNMP/TLS/TCP appears to be a viable option for network management.