KubeVirt v1.8 for Kubernetes VMs

A lot of platform teams still have workloads that cannot move cleanly into containers. That is where KubeVirt matters. It lets operators manage virtual machines with Kubernetes APIs, so VM-based services can live beside container workloads without building a separate control plane. The new v1.8 release is worth a look because it improves scale, adds more flexibility around hypervisors, and sharpens support for security-sensitive workloads.

What is KubeVirt?

KubeVirt is an open source virtualization layer for Kubernetes. It extends the cluster with CRDs and controllers that let teams create, run, and manage virtual machines using familiar Kubernetes patterns. That makes it useful for gradual modernization, legacy app hosting, edge deployments, and mixed VM plus container platforms.

The v1.8 release aligns with Kubernetes v1.35 and brings a set of updates that matter for SRE and platform engineering teams running real infrastructure.

Key Features

• Hypervisor Abstraction Layer: KubeVirt introduces a new Hypervisor Abstraction Layer, which opens the door to multiple hypervisor backends beyond the default KVM path.
• Confidential computing improvements: Intel TDX attestation support helps confidential VMs prove they are running on confidential hardware.
• Better VM networking: The passt binding is now a core binding, and live updates to NAD references reduce disruption when changing a VM's backing network.
• Incremental backups with CBT: Changed Block Tracking enables storage-agnostic incremental VM backups, which cuts backup windows and reduces storage and network overhead.
• Scale and performance work: The community reports successful performance testing up to 8000 VMIs, which is a useful signal for larger control-plane deployments.

Installation

KubeVirt can be installed on a Kubernetes cluster with the published operator manifests:

export RELEASE=v1.8.0
kubectl apply -f https://github.com/kubevirt/kubevirt/releases/download/${RELEASE}/kubevirt-operator.yaml
kubectl apply -f https://github.com/kubevirt/kubevirt/releases/download/${RELEASE}/kubevirt-cr.yaml

After that, check the control plane components:

kubectl get kubevirt -n kubevirt
kubectl get pods -n kubevirt

Usage

A simple path is to install KubeVirt, apply a VM manifest, and manage the guest with standard Kubernetes workflows:

kubectl apply -f vm.yaml
kubectl get vm
kubectl get vmi
virtctl start <vm-name>

For SRE teams, the real value is not just that VMs run on Kubernetes. It is that VM lifecycle, policy, observability, and automation can now fit into the same GitOps and platform engineering workflows used for containers.

Operational Tips

Start by identifying VM workloads that need Kubernetes-adjacent automation, not full containerization. Backups are a good place to test v1.8 because incremental backup support can lower both cost and operational drag. If you run regulated or high-sensitivity workloads, the new confidential computing work is another strong reason to evaluate the release. For larger deployments, keep an eye on controller memory use as VMI counts rise and test your own cluster profile before broad rollout.

Conclusion

KubeVirt v1.8 pushes Kubernetes virtualization forward in a practical way. Hypervisor flexibility, stronger confidential VM support, better networking, and incremental backups all make the release relevant for teams trying to unify mixed infrastructure under one platform.

Looking to automate infrastructure operations? Akmatori helps SRE teams reduce toil with AI agents built for real production workflows. For reliable global infrastructure, check out Gcore.