SLA Violation Prediction In Cloud Computing: A Machine Learning Perspective

TL;DR

This paper investigates machine learning techniques, specifically Naive Bayes and Random Forests, to predict rare SLA violations in cloud computing, addressing class imbalance with re-sampling methods and achieving high accuracy.

Contribution

It demonstrates that Random Forests combined with SMOTE-ENN re-sampling significantly improve SLA violation prediction accuracy in cloud systems.

Findings

Random Forest with SMOTE-ENN achieves 99.88% accuracy.

Re-sampling methods effectively handle class imbalance.

High F1 score of 0.9980 indicates reliable predictions.

Abstract

Service level agreement (SLA) is an essential part of cloud systems to ensure maximum availability of services for customers. With a violation of SLA, the provider has to pay penalties. In this paper, we explore two machine learning models: Naive Bayes and Random Forest Classifiers to predict SLA violations. Since SLA violations are a rare event in the real world (~0.2 %), the classification task becomes more challenging. In order to overcome these challenges, we use several re-sampling methods. We find that random forests with SMOTE-ENN re-sampling have the best performance among other methods with the accuracy of 99.88 % and F_1 score of 0.9980.