Design of Visual Log Analysis System

The log file can record the detailed operating information of the operating system and the running details of various service programs. Developers can use it to analyze system performance, check system exceptions and errors, and optimize the system at any time. Therefore, the importance of the log in the system’s development and maintenance is unquestionable. However, with the increasing complexity of modern software systems, the log size is also increasing rapidly. Hence, traditional log analysis methods that rely mainly on manually customized regular expressions are not enough. A series of automated log analysis tools have emerged. This article designed a visual log analysis system called VLAS (Visual Log Analysis System). This paper first introduces log analysis’s research status at home and abroad, as well as the evaluation of existing log analysis tools on various data sets, then briefly describes the relevant theories of log analysis. Then it introduces the design of the system from the overall architecture to the specific modules, and finally shows the functions that have been implemented in the system. VLAS can efficiently and quickly parse out structured log files and event template files that aggregate event statistics from the original log text. The structured log files are used for subsequent anomaly detection.The system implements three functions containing log parsing, anomaly detection and visual display. It integrates data processing and result analysis. Users can select appropriate algorithms according to the log data from different systems and get a better user experience.