Which factor triggers autoscaling on CloudHub?

Autoscaling on CloudHub is primarily triggered by CPU or memory usage thresholds. This mechanism ensures that applications can automatically adjust their resources in response to actual performance needs, maintaining optimal performance and cost efficiency.

When the resource utilization of an application, particularly in terms of CPU or memory consumption, reaches a predefined threshold, CloudHub automatically scales the number of application workers up or down. This adjustment helps prevent application slowdowns during peak usage and ensures resources are not wasted during low-demand periods, thereby improving both responsiveness and cost management for running applications.

Time-based policies, while useful for planning and forecasting resource requirements, do not directly trigger autoscaling based on real-time application performance. User-defined scaling requests and application version changes also do not trigger automatic scaling directly; instead, they may be part of a broader resource management or deployment strategy but do not react the same way to usage patterns as CPU or memory thresholds do.