Support Vector regression for Service Level Agreement violation prediction
Ahmad Fadzil Mohamad Hani, Irving Vitra Paputungan, Mohd Fadzil Hassan · 2013
SLA is a contract between service providers and consumers, mandating specific numerical target values which the service needs to achieve. For service providers, preventing SLA violation becomes very important to enhance customer trust and avoid penalty charging. Therefore, it is necessary for providers to forecast possible violations as much as possible before they actually happen. Time series analysis based on Support Vector Machine for regression is proposed for predicting SLA violations. It will analyse historical data of performance to provide estimated upcoming data. A validation using 120 days sample data shows that Support Vector Machine could predict service performance data in cloud database. The prediction accuracy is considerably high in this particular case; it is more than 80%.