A log parsing method based on non-fixed search tree
Yinghui Xu, Yanqing Wang, Ruiqiang Guan · 2024
Log data records important information about the system during operation, and plays an important role in system anomaly detection. As the size of logs continues to increase, efficiently processing logs has become an important issue. In this paper, we propose a log parsing method called CUD, which is based on a non-fixed search tree. The method encodes log parsing rules in a structural tree format and parses log messages, and is tested on four publicly available log datasets. Experimental results show that compared to other online methods, CUD has higher accuracy and achieves efficient and accurate parsing of log data.