Periodicity analysis on heterogeneous protocols

摘要

Disclosed are systems and methods for detecting behavior of periodic events from machine generated logging by: collecting heterogeneous log messages, each log message containing a timestamp and text content with one or more fields; Recognizing log formats from log messages; Transforming the textual content into a set of time series data, a time series for each protocol format; Analyzing the set of time series data and building a category model for each type of periodic event in heterogeneous protocols during a training phase; and applying the category model to a stream of time-series data of heterogeneous live log messages and generating a flag at a time-series data point that violates the category model and generating an alarm report for the corresponding log message during live operation.