AI That Predicts Server Demand Could Mean Fewer Crashes, Lower Prices

Researchers benchmarked how workflow topology affects task-level resource intensity prediction for large-scale cloud workflows. These workflows are often structured as directed acyclic graphs (DAGs), where under-provisioning can cause critical bottlenecks and over-provisioning leads to unnecessary costs — which is why accurate, task-level prediction of resource intensity like CPU load and memory usage matters. The central question: to what extent does what part of the DAG topology influence task-level resource intensity, and what is the most effective way to model this influence? Their finding: topology is a critical feature for accurate prediction. Models incorporating important topological information — even through simple handcrafted features — significantly outperform baseline models. Graph-native models provide the highest accuracy, achieving low mean absolute errors for both CPU and memory predictions, and can still be combined with simple topological features that they do not learn for better performance.