一句话：提出样本高效的扩散模型 Diffusion+，对云系统时序缺失数据进行插补，以提升下游故障预测性能。
大规模云系统（如 Microsoft 365）中的磁盘、节点故障会威胁服务可靠性，造成服务中断与经济损失。已有工作多聚焦于故障预测并提前采取行动，但其模型训练与预测所依赖的数据常存在缺失，数据质量差限制了预测性能。本文将问题定位为通过数据插补提升数据质量，提出 Diffusion+：一种样本高效的扩散模型，以已观测数据为条件高效地插补缺失值。作者在工业数据集与实际上线应用中进行了评估，结果表明该模型能够提升下游故障预测任务的性能，说明数据插补是改善云故障预测的有效途径。
Ensuring reliability in large-scale cloud systems like Microsoft 365 is crucial. Cloud failures, such as disk and node failure, threaten service reliability, causing service interruptions and financial loss. Existing works focus on failure prediction and proactively taking action before failures happen. However, they suffer from poor data quality, like data missing in model training and prediction, which limits performance. In this paper, we focus on enhancing data quality through data imputation by the proposed Diffusion+, a sample-efficient diffusion model, to impute the missing data efficiently conditioned on the observed data. Experiments with industrial datasets and application practice show that our model contributes to improving the performance of downstream failure prediction.
@inproceedings{YangYWL0LW0LBRL23,
title = {Diffusion-Based Time Series Data Imputation for Cloud Failure Prediction at Microsoft 365},
author = {Fangkai Yang and Wenjie Yin and Lu Wang and Tianci Li and Pu Zhao and Bo Liu and Paul Wang and Bo Qiao and Yudong Liu and Mårten Björkman and Saravan Rajmohan and Qingwei Lin and Dongmei Zhang},
booktitle = {FSE},
year = {2023},
doi = {10.1145/3611643.3613866}
}