Kubernetes v1.36: In-Place Pod-Level Resources Vertical Scaling Graduates to Beta
The Kubernetes community has announced that In-Place Pod-Level Resources Vertical Scaling has officially graduated to Beta status in version 1.36. This feature, enabled by default via the InPlacePodLevelResourcesVerticalScaling feature gate, allows users to update the aggregate resource budget for running Pods without necessarily restarting containers. It builds upon previous releases where Pod-Level Resources reached Beta in v1.34 and In-Place Pod Vertical Scaling achieved General Availability in v1.35. The update simplifies management for complex Pods, such as those with sidecars, by allowing containers to share a collective resource pool that can be adjusted on-the-fly. When resizing, the Kubelet checks individual container resize policies to determine if a dynamic cgroup update is possible or if a restart is required. This capability is particularly beneficial for handling peak demand scenarios, as it eliminates the need for manual per-container recalculations when expanding shared resource limits. The announcement highlights significant progress in Kubernetes' ability to handle resource efficiency and operational flexibility for modern cloud-native applications.
Wire timeline
Kubernetes v1.36: In-Place Pod-Level Resources Vertical Scaling Graduates to Beta
The Kubernetes community has announced that In-Place Pod-Level Resources Vertical Scaling has officially graduated to Beta status in version 1.36. This feature, enabled by default via the InPlacePodLevelResourcesVerticalScaling feature gate, allows users to update the aggregate resource budget for running Pods without necessarily restarting containers. It builds upon previous releases where Pod-Level Resources reached Beta in v1.34 and In-Place Pod Vertical Scaling achieved General Availability in v1.35. The update simplifies management for complex Pods, such as those with sidecars, by allowing containers to share a collective resource pool that can be adjusted on-the-fly. When resizing, the Kubelet checks individual container resize policies to determine if a dynamic cgroup update is possible or if a restart is required. This capability is particularly beneficial for handling peak demand scenarios, as it eliminates the need for manual per-container recalculations when expanding shared resource limits. The announcement highlights significant progress in Kubernetes' ability to handle resource efficiency and operational flexibility for modern cloud-native applications.
Kubernetes Blog