[ANNOUNCE] dra-driver-cpu v0.2.0 is released

[Francesco Romani]

Hello everyone,

We are pleased to announce the release of dra-driver-cpu v0.2.0.

dra-driver-cpu is a Dynamic Resource Allocation (DRA) driver for CPU resources.
It enables Kubernetes clusters to manage and assign exclusive CPUs to
workloads using the DRA framework.
The driver replaces the CPUManager functionalities implemented in the kubelet.

Release Notes and Download:
https://github.com/kubernetes-sigs/dra-driver-cpu/releases/tag/v0.2.0

Requirements: Kubernetes 1.36.0+ (or 1.34.0+ with the
DRAConsumableCapacity alpha feature gate enabled),
Helm v3 for Helm-based deployment.
Please see the documentation and the release notes for the
configuration requirements.

Release hot topics:
- Helm Chart Deployment: Helm is now the recommended installation
method, replacing the `install.yaml` shipped in v0.1.0.
`install.yaml` is still included for backward compatibility but will
be removed in the next release.
- PCIe Root Topology Alignment: The driver can now optionally discover
PCIe root topology and expose standard
`resource.kubernetes.io/pcieRoot` device attributes, enabling
schedulers to co-locate CPU allocations with GPUs or NICs
sharing the same PCIe root. Requires the alpha
`DRAListTypeAttributes` feature gate to be enabled in the cluster.
- ARM64 Node Support: Enabling multi-arch image builds (amd64 and
arm64). CPU topology discovery has been refactored to
strictly use sysfs and fixes have been implemented for L3 cache
discovery, SMT detection, and NUMA affinity masking on
ARM64 hardware.
- Machine Grouping Mode: Introduced a new `--group-by=machine`
configuration. In this mode, the driver exposes a single
node-wide capacity device and enforces exact CPU assignments
provided via the claim's opaque configuration parameters.
This is being evaluated as a replacement path for the individual
mode when using external schedulers to enforce precise
CPU allocation.
- Reliability and Atomicity: Hardened allocation and teardown
ordering, NRI restart recovery, idempotent setup,
and standardized CDI configuration management.
- Debugging Helper (dracpu-gatherinfo): Added a built-in helper tool
to scan and print the host node's hardware topology
in the driver's internal format, simplifying the troubleshooting of
the driver allocation issues.

Please check the release notes for full highlights.

Huge thanks to everyone who contributed to this release:
@AutuSnow, @back1ash, @ffromani, @fmuyassarov, @gauravgahlot,
@Karthik-K-N, @pravk03, @rocker-zhang

Kudos to our new contributors: @back1ash, @Karthik-K-N, @gauravgahlot,
@rocker-zhang

Feedback and contributions are always welcome. Please feel free to
reach out on kubernetes slack in #wg-device-management
or file issues on the GitHub repository.

Best regards,

Francesco Romani
On behalf of the dra-driver-cpu maintainers

--
Francesco Romani -- software engineer @ Red Hat
https://github.com/ffromani