TL;DR

Kubernetes 1.35 introduces the In-Place Pod Resize feature, allowing dynamic adjustment of CPU and memory resources within running pods without restarts.

What happened

The In-Place Pod Resize feature in Kubernetes has reached stable (GA) status after its initial introduction as alpha in v1.27 and beta in v1.33. This allows for the modification of CPU and memory limits on running pods, enhancing resource efficiency and workload flexibility.

Why it matters for ops

In-place pod resizing reduces operational disruptions by enabling the adjustment of resources without downtime or loss of state, making it ideal for workloads sensitive to latency and restarts.

Action items

  • Review official Kubernetes documentation for implementation details
  • Consider integrating with autoscalers like VPA for enhanced efficiency
  • Monitor new Kubelet metrics for better observability

Source link

https://kubernetes.io/blog/2025/12/19/kubernetes-v1-35-in-place-pod-resize-ga/