RedHat OpenShift Virtualization [Part 1] – The bridge between VMs and Containers

The IT infrastructure of many companies is caught between two conflicting priorities: on one hand, business-critical applications have been running stably in virtual machines (VMs) for years. On the other hand, modern developments, microservices and agile teams require the use of containers and Kubernetes. This is exactly where RedHat comes in. With RedHat OpenShift Virtualization (RHOCPV), VMs become first-class citizens in Kubernetes and OpenShift. The technology is based on KubeVirt, an open-source project, and enhances RedHat OpenShift with the ability to seamlessly manage virtual machines in addition to containers. The result: administrators and developers get a unified platform on which workloads – whether containers or VMs – are operated with the same processes for deployment, scaling, monitoring, and security.

Competitors: VMware, Proxmox & Co. 

Before understanding why OpenShift Virtualization is exciting, it is worth taking a look at the alternatives: 

• VMware vSphere/ESXi: The previous de-facto standard in many datacenters. VMware offers a mature virtualization platform, a huge range of functions, and broad ecosystem integration. However, since the acquisition of Broadcom, the costs are high and container integration is only possible via additional products such as Tanzu.  

• Proxmox VE: An open source project that combines virtualization (KVM/QEMU) and containers (LXC).Proxmox scores points for its simplicity and cost-free nature, but is less common in enterprise environments and has less focus on cloud-native scenarios compared to OpenShift.

• Other providers such as Microsoft Hyper-V have the ability to run containers, albeit in a very limited way. Since Windows cannot natively run Linux containers, Hyper-V first has to start a Linux VM in which the containers run.

So the real question is: Do companies really still need a separate virtualization platform when Kubernetes is already the strategic target platform? This is exactly where RHOCPV plays to its strengths.

Why OpenShift Virtualization? 

1. Consolidation of platforms 
Instead of running two worlds in parallel- containers in Kubernetes and VMs on VMware/Proxmox- companies can simplify their infrastructure. Everything runs in the same cluster, with the same tools for logging, monitoring, security, and deployment.

2. Modernization without a big bang 
Not every legacy application can be containerized immediately. With RHOCPV, these applications can continue to run in VMs, while new components run in containers. This results in gradual modernization.

3. Cost Efficiency 
License costs for proprietary virtualization solutions such as VMware can be significant. Since RHOCPV is based on OpenShift, existing investments in the container platform are utilized- without additional ‘silo costs.’

4. DevOps & CI/CD for VMs 
VMs benefit from the same automation and deployment pipelines as containers. Templates, infrastructure as code and GitOps can be applied to VMs in exactly the same way.

5. Unified Storage
Since the storage backend is configured in OpenShift, it is available uniformly in both containerized and VM instances.

Why network connectivity is the key issue

Integrating VMs into Kubernetes sounds appealing- but one area is crucial to success: the network.

While containers typically work with service-oriented communication patterns (service mesh, load balancer, DNS-based discovery), VMs often rely on classic network concepts: static IPs, VLANs, legacy protocols or appliances that require a ‘real’ Layer 2 network.

The challenges here are:  

• Compatibility: Legacy applications in VMs expect to function as they would in a traditional VM environment. For example, network interfaces must not suddenly disappear in an overlay structure.

• Performance: Virtual machines that run data-intensive workloads (e.g. databases or storage getaways) often require near-bare-metal performance in the network. This is where SR-IOV (Single Root I/O Virtualization) comes into play, which can also be supported in RHOCPV.

• Integration into the Kubernetes network model: VMs must not only communicate with classic systems, but also with containers in the cluster. This requires flexible but consistent network concepts.

Especially in hybrid scenarios- legacy VMs communicating with modern microservices- it becomes clear that nothing works without a clean network communication.

Conclusion: Introducing a new way of thinking about infrastructure

OpenShift virtualization is more than just another virtualization solution. It bridges two worlds: stable, proven VMs and modern, containerized applications. Companies that currently run both traditional workloads and new cloud-native services can use it to consolidate their infrastructure, save costs and accelerate their modernization efforts. But one thing is also clear: the path to VMs in Kubernetes stands and falls with the network. Only when legacy VMs can interact with container workloads naturally and efficiently will OpenShift virtualization unfold its full potential.

This lays the foundation- in the coming articles, we will delve deeper into architecture, network options, storage integration, and practical examples from real-world applications.

zurück