- DRA: scheduling pods is up to 16x faster, depending on the scenario. Scheduling throughput depends a lot on cluster utilization. It is higher for lightly loaded clusters with free resources and gets lower when the cluster utilization increases. ([#127277](https://github.com/kubernetes/kubernetes/pull/127277),...

- DRA: Device taints enable DRA drivers or admins to mark device as unusable, which prevents allocating them. Pods may also get evicted at runtime if a device becomes unusable, depending on the severity of the taint and whether the claim tolerates the taint. ([#130447](https://github.com/kubernetes/kubernetes/pull/130447), [@pohly](https://github.com/pohly)) [SIG API Machinery, Apps, Architecture, Auth, Etcd, Instrumentation, Node,...

* [action required] The WATCHLIST calls are now reported as WATCH verbs in prometheus for the apiserver_request_* series.  A new "scope" label is added to all apiserver_request_* values that is either 'cluster', 'resource', or 'namespace' depending on which level the query is performed at.([#52237](https://github.com/kubernetes/kubernetes/pull/52237))


#### Go Client

* Add support for adding/removing finalizer depending on PVC protection feature gate ([#61370](https://github.com/kubernetes/kubernetes/pull/61370), [@gnufied](https://github.com/gnufied))

        or provides additional installation flags.
    *   This dependency affected anyone installing TensorFlow 2.15 alongside
        NVIDIA CUDA dependencies via `pip install tensorflow[and-cuda]`.
    *   Depending on the installation method, TensorFlow 2.14 would be installed
        instead of 2.15, or users could receive an installation error due to
        those missing dependencies.

  package for one more release, but in you must adapt your scripts to install `crictl` manually from
  https://github.com/kubernetes-sigs/cri-tools/releases or a different location.
  
  The `kubeadm` package will stop depending on the `cri-tools` package in Kubernetes 1.32, which means that

element\n */\nconst getNextActiveElement = (list, activeElement, shouldGetNext, isCycleAllowed) => {\n  const listLength = list.length\n  let index = list.indexOf(activeElement)\n\n  // if the element does not exist in the list return an element\n  // depending on the direction and if cycle is allowed\n  if (index === -1) {\n    return !shouldGetNext && isCycleAllowed ? list[listLength - 1] : list[0]\n  }\n\n  index += shouldGetNext ? 1 : -1\n\n  if (isCycleAllowed) {\n    index = (index + listLength) %...

  When the scheduling queue receives a cluster event, before moving each Pod from unschedulable pod pool to activeQ/backoffQ, it will call QueueingHintFn of plugins that rejected each Pod in the previous scheduling cycle.
  Depending on the value returned from QueueingHintFn, the scheduling queue changes how it queues each Pod:
  - if more than one QueueingHintFn returns QueueImmediately, it queues Pod to activeQ.

- Kubectl cluster-info dump --output-directory=xxx now generates files with an extension depending on the output format. ([#82070](https://github.com/kubernetes/kubernetes/pull/82070), [@olivierlemasle](https://github.com/olivierlemasle)) [SIG CLI]

with two steps: (1) assert copyright on the software, and (2) offer
you this License which gives you legal permission to copy, distribute
and/or modify the software.

  A secondary benefit of defending all users' freedom is that
improvements made in alternate versions of the program, if they
receive widespread use, become available for other developers to
incorporate.  Many developers of free software are heartened and