A METHOD FOR THE STATE EVALUATION OF DISTRIBUTED INFORMATION SYSTEM

Abstract

Information system has many geographically distributed objects and a administrator can’t evaluate its current state or he must process too many indicators of state and this is not always possible. Article describes the method of evaluation a state the distributed information system that has realized with elements of the artificial intelligence. Artificial neural networks may be used for clustering and generalization of data and as a associative memory. This enables to minimize the time required for computing and search the solutions to emerging problems. This method is realized as the model that uses the processed data received from controlled objects of the distributed information system. The model calculates generalized indicator that determines the state of distributed information system as one value that is easy ranked. Because of territorial width and large numbers of objects, the administrator does not always have the opportunity to assess the current state of the information system. Processing a large number of values of different indicators is necessary for such an evaluation. The article describes a way to assess the current state of the distributed information system that is implemented using elements of artificial intelligence. The use of artificial neural networks for clustering and aggregation of the data, as well as associative memory, can minimize the time required for the calculations and searching solutions for emerging problems. The proposed method is implemented as a model, which has the source data that are processed by the data collected from sites of the distributed information system. The model calculates the common ratio that defines the current state of the distributed information system in a single and easy calculated value.