TL;DR
A novel explainable deep time-series-based framework for real-time load forecasting in cloud environment, incorporating explainability techniques (XAI) such as SHapley Additive exPlanations (SHAP), which demonstrates a higher level of accuracy in the prediction of workload.
A dynamic recurrent neural network is proposed to accurately predict workloads and integrates an auto-encoder to effectively extract representations from the original workload data with high dimensionality to enable adaptive and accurate predictions for highly variable workloads.
Experimental results demonstrated that the forecasting–detection pipeline effectively adapts to varying consumption behaviors, yielding improved forecasting accuracy and more reliable AD, particularly over longer forecasting horizons, confirming its suitability for practical smart-grid deployment.
The proposed MST-iTransformer model achieves simultaneous reductions in prediction errors and inference latency, validating the effectiveness of the multi-scale temporal encoding, sparse attention mechanism, and adaptive feature selection modules in improving forecasting accuracy and computational efficiency.
A deep learning-based failure prediction model that integrates Convolutional Neural Networks and Bidirectional Long Short-Term Memory networks to identify job failures before they occur is presented, improving the performance of cloud computing applications by reducing job failures and optimising resource utilisation.
A Hybrid Temporal-Feature Attention enhanced Multi-task Stacking model (HTAM-Stack) for multivariate cluster workload forecasting, designed to jointly capture temporal dependencies and cross-resource feature interactions, achieves competitive prediction performance under complex and dynamic workload conditions.
We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.
