Serverless computing offers automatic resource management and pay-per-use execution, but effective autoscaling remains challenging due to dynamic workloads, cold-start latency, and dependencies among functions.
We present a dependency-aware autoscaling framework that integrates graph-based bottleneck identification, short-term workload forecasting, multi-model consensus, and cost-aware scaling control.
Serverless applications are represented as directed dependency graphs, with structurally important functions identified using weighted degree centrality. Resource demand is predicted using lightweight MLP, LSTM, and CNN models. Their outputs are combined through a performance-weighted probabilistic ensemble inspired by Bayesian model averaging. The controller incorporates cold-start awareness and cost comparison to select among scale-up, scale-down, and hold actions.
Experiments using real workload traces show that supervised forecasting significantly outperforms unsupervised clustering for autoscaling decision generation. The proposed ensemble achieves 99.88% prediction accuracy and reduces prediction error compared to representative hybrid forecasting methods. Evaluations across multiple cloud pricing models demonstrate consistent infrastructure cost reductions while maintaining performance targets.
The results indicate that combining dependency analysis, multi-expert forecasting, and cost-aware control provides a robust and practical solution for serverless autoscaling.
Blogger's Review: This research presents an innovative solution to the autoscaling challenge in serverless computing by integrating various models and dependency analysis. Its practical value lies in effectively managing dynamic workloads and costs, showcasing significant efficiency in resource management through high-accuracy prediction models. This approach merits promotion in real-world cloud services.