Abnormality detection device, abnormality detection system, and abnormality detection method

Abstract

To provide an abnormality detection method that can detect abnormality in micro-service with high accuracy.SOLUTION: An abnormality detection device includes: a telemetry information acquisition unit that acquires first telemetry information including log information, trace information, and metric information from a first set of micro-service; an embedding generation unit that generates trace embedding corresponding to the trace information, generates log embedding corresponding to the log information, and generates metric embedding corresponding to the metric information; a model generation unit that generates, on the basis of the trace embedding, the log embedding, and the metric embedding, a machine learning model for abnormality detection that determines presence or absence of abnormality in the micro-service; and an abnormality detection unit that analyzes, using the machine learning model for abnormality detection, second telemetry information acquired for a second set of micro-service and then generates an abnormality detection result indicating presence or absence of abnormality in the second micro-service.SELECTED DRAWING: Figure 2

Images