Our organization is reassessing its virtualization platform. Nutanix is no longer being considered because its renewal pricing was higher than VMware in several comparable cases, and management has rejected other alternatives as insufficiently mature.
The main constraints are:
- The team has little Kubernetes experience.
- VMware would cost roughly four times as much as OpenShift.
- About 95% of our environment consists of virtual machines.
- We need disaster recovery, and VMware Site Recovery Manager would be convenient.
- Around 65% of our existing hardware is not supported by the newer VMware version we would need.
- Approximately 75% of the infrastructure is Linux-based, including Oracle databases and RHEL application servers.
- Management is reluctant to adopt Kubernetes, even though it could potentially reduce resource requirements.
- Many of our applications are web-based.
I'm trying to determine whether OpenShift is genuinely a practical VMware replacement for this environment, or whether we should focus on another hypervisor and plan a hardware refresh. Any advice on the architecture, migration effort, and realistic alternatives would be appreciated.
5 Answers
Before choosing based only on the headline price, get a complete three-year comparison. Include licenses, support, migration work, hardware replacement, backup and DR tooling, training, consultants, and additional staffing. Also, unsupported hardware is a problem regardless of the platform you select, so a refresh plan is unavoidable. If VMware’s current offer is competitive after negotiation, staying may be less disruptive than changing platforms solely to reduce licensing costs.
The management decision is the biggest constraint here. If alternatives such as Proxmox or XCP-ng are automatically rejected, and Kubernetes is also considered unacceptable, the practical choices are already limited. Build a short business case with a current-state inventory, supported-hardware matrix, five-year costs, migration risks, and staffing requirements. That gives leadership a clear choice between renewing VMware, funding a hardware refresh and hypervisor migration, or investing in Kubernetes skills for OpenShift.
For a straightforward hypervisor replacement, investigate Proxmox seriously. Several administrators report running large clusters reliably, and the platform can be much less expensive than VMware. Put the savings toward replacing unsupported hardware and training the team. There is still a learning curve, and you would need to validate backup, storage, networking, monitoring, and disaster recovery before committing, but it fits your mostly-VM workload much better than OpenShift.
XCP-ng is another hypervisor option worth evaluating for Linux-heavy workloads. It is closer to VMware in architecture and operational model than OpenShift, so the transition may be less disruptive. You should still run a proof of concept covering Oracle databases, RHEL application servers, live migration, storage performance, backups, and DR. Do not approve a platform based on a feature checklist alone.
OpenShift and vSphere are not direct equivalents. VMware is primarily a virtualization platform, while OpenShift is a Kubernetes platform that can also run virtual machines through its virtualization features. Since your environment is 95% VMs and the team is not comfortable with Kubernetes, moving to OpenShift would be a major operational and skills transition, not simply a cheaper VMware replacement. It makes the most sense if you also have a serious plan to modernize some applications into containers.
That is my concern too. The pricing looks attractive, but I’m not sure management understands that OpenShift still requires Kubernetes expertise.
Proxmox was suggested, and the team already has some experience with it, but the CTO rejected it despite the pricing and technical fit.