The in-place pod vertical scaling feature is managed through a feature gate, and its availability depends on both the Kubernetes version supported by AKS and the feature's stability for production workloads.
In-Place Vertical Pod Scaling was introduced in Kubernetes 1.27, allowing for dynamic adjustments of CPU and memory resources without recreating pods. However, this feature is still in its alpha stage and requires a feature gate to be enabled before use.
At this time, there is no official release time for when specific feature gates like in-place pod vertical scaling will be enabled, as Microsoft and the Kubernetes community typically release features gradually, following extensive testing and a stable rollout process.
To stay updated on its progress:
- Monitor Microsoft Learn and AKS documentation for future announcements.
- Follow ongoing discussions and testing efforts on GitHub: https://github.com/kubernetes/autoscaler/issues/4016 https://github.com/kubernetes/autoscaler/pull/7673
Hope this helps!
Please Let me know if you have any queries.