Failure cause identification system, failure cause identification method and failure cause identification program

摘要

PROBLEM TO BE SOLVED: To improve the efficiency of operation of an ICT infrastructure by shortening the time required to identify the cause of a failure. SOLUTION: A failure cause identification system 150 causes an abnormality in an application from the configuration information of a network system 100 based on a first concept (“when a plurality of abnormalities occur at the same time, a common factor exists”). Extract the related nodes that are related to the new node. Each metric of the extracted related node is set as an "objective variable", and the others are candidates for "explanatory variable", and an explanatory variable that can be used as a prediction model for each objective variable is selected from the candidates. An abnormality of the objective variable is detected by JIT determination, and the objective variable for which the abnormality is detected and its explanatory variable are detected. Next, based on the second idea (“If there is a factor common to multiple anomalies, it is highly likely that it is the cause of the anomaly”), the objective common to each explanatory variable of the objective variable in which the anomaly is detected. The number of variables is extracted, and the one with the largest number is the first candidate for the cause of abnormality. [Selection diagram] Fig. 1