Building and Validating Deep Learning Models for Forecasting the Quality of Cloud Services

Context: Cloud services operate in highly dynamic and heterogeneous environments, requiring a continuous and accurate assessment of service quality. While Quality of Service (QoS) models are widely used to monitor performance, deep learning (DL) architectures—such as Long Short-Term Memory (LSTM) and Bidirectional Gated Recurrent Units (BI-GRU)—offer enhanced capabilities for forecasting potential SLA violations. However, many existing experiments in this domain suffer from methodological shortcomings, including the use of outdated or proprietary datasets, a narrow set of QoS metrics, incomplete documentation of model architectures and training procedures, and a lack of statistical rigor, all of which undermine reproducibility and applicability in industrial contexts. Objective: This study empirically compares the performance of BI-GRU, LSTM, and AutoRegressive Integrated Moving Average (ARIMA) models for QoS forecasting, using a rigorously designed experimental protocol that addresses these limitations. Method : We developed a multi-metric QoS dataset covering five months of operational data from a cloud service in an IT company, comprising 16 QoS metrics. Forecasting models were trained and evaluated using Root Mean Squared Error (RMSE), Mean Absolute Error (MAE), and Mean Absolute Percentage Error (MAPE), with training time as an efficiency indicator. Results: BI-GRU outperformed ARIMA across all QoS metrics models and outperformed LSTM in 10 out of 16 QoS forecasting models (62.5%) with competitive or shorter training times. Conclusion: Our findings demonstrate that BI-GRU models deliver superior accuracy and efficiency and that methodological rigor supports their applicability for proactive QoS management and informed decision-making in industrial cloud service environments.