English

Why does Prediction Accuracy Decrease over Time? Uncertain Positive Learning for Cloud Failure Prediction

Distributed, Parallel, and Cluster Computing 2024-02-02 v1 Artificial Intelligence

Abstract

With the rapid growth of cloud computing, a variety of software services have been deployed in the cloud. To ensure the reliability of cloud services, prior studies focus on failure instance (disk, node, and switch, etc.) prediction. Once the output of prediction is positive, mitigation actions are taken to rapidly resolve the underlying failure. According to our real-world practice in Microsoft Azure, we find that the prediction accuracy may decrease by about 9% after retraining the models. Considering that the mitigation actions may result in uncertain positive instances since they cannot be verified after mitigation, which may introduce more noise while updating the prediction model. To the best of our knowledge, we are the first to identify this Uncertain Positive Learning (UPLearning) issue in the real-world cloud failure prediction scenario. To tackle this problem, we design an Uncertain Positive Learning Risk Estimator (Uptake) approach. Using two real-world datasets of disk failure prediction and conducting node prediction experiments in Microsoft Azure, which is a top-tier cloud provider that serves millions of users, we demonstrate Uptake can significantly improve the failure prediction accuracy by 5% on average.

Keywords

Cite

@article{arxiv.2402.00034,
  title  = {Why does Prediction Accuracy Decrease over Time? Uncertain Positive Learning for Cloud Failure Prediction},
  author = {Haozhe Li and Minghua Ma and Yudong Liu and Pu Zhao and Lingling Zheng and Ze Li and Yingnong Dang and Murali Chintalapati and Saravan Rajmohan and Qingwei Lin and Dongmei Zhang},
  journal= {arXiv preprint arXiv:2402.00034},
  year   = {2024}
}
R2 v1 2026-06-28T14:33:34.843Z