الفهرس 
abstractOnly 14 pages are availabe for public view
 |  | from | 171 |  |  |
| | | from | 171 | | |
Abstract
In the design of a computer network, one of the fundamental considerations is the reliability and availability of the communication ~aths between all pairs of nodes in the network. These characteristics are strongly dependent on the topological layout of the communication links in addition to the reliability and availability of the individual computer systems and communication facilities. This research introduces a proposed approach to analyze the network reliability under some performance indices constraints by the use of artificial intelligence techniques. The .thesis aims to develop a knowledge-based system that helps in designing large scale computer networks with the objectives of minimization of total network probability of failure, delay and cost of the network links. The proposed system offers the following capabilities It Establishes and builds a database for the characteristics and attributes of the network under investigation allowing dynamic adaptation to any changes in the network data such as addition and/or deletion of nodes and/or links. It provides a complete visibility about the network under study by showing the network topology line graph and displaying the network data and attributes. It performs knowledge acquisition to have information that reflects the behaviour of the network and performance attributes at each node of the network, such as network reliability between source node and all nodes in the network, node degree and flow information at each node, beside a complete information about optimum path attributes between the source node and any other node in the network. It provides an expert system that can be consulted to show graphs and display information about the network perform~nce attributes through the following two analysis modes: First mode is Network Reliability analyzer that shows graphs providing information, at each node, about network reliability, node degree and traffic flow. Moreover, it provides capability to perform sensitivity analysis of the network reliability attribute, for single network topology and/or multiple topologies. Second mode is Optimum path analyzer which displays and shows complete information about optimum path attributes between the source node and all other nodes• in the network. This mode perL~rms determination of optimum path performance for two ordered attributes, e.g. the most reliable path with minimum number of hops. During the different phases of developing the proposed system the modern techniques of system analysis and design have been used. AI techniques, e.g. Problem-reduction technique, and OR techniques , such as OP, have been applied, beside the use of ROBS to build and maintain both the database and knowledge base for the network under study. As will be clear through the material of this thesis, it is useful to use the proposed system to obtain network performance analysis with the focussing on the network reliability attribute.