ml-SFP: System Failure Prediction Method Based on Machine Learning

As system reliability and availability become the key driving factors in various IT services, the ability of predicting system failure is considered one of the most important aspects to avoid data and revenue loss. In this paper, we propose a machine learning-based system failure prediction method, called machine learning-based System Failure Prediction (ml-SFP), which automatically enables to identify the optimal machine learning model and its hyper-parameter values generating better performance, without requiring a deep knowledge of system failure prediction. We present the performance evaluation of ml-SFP to verify its effectiveness in predicting failures.