Service

Move off VMware, onto a platform you choose

We migrate virtualization estates off VMware ESXi as one programme: the hypervisor, the storage and the network underneath. The target is chosen to your constraints, open source or licensed, and where it falls short, we extend it in code. On one programme, 10,000+ VMs moved to Proxmox and Ceph with a $0 hypervisor licence fee.

Measured impact

Why teams are leaving VMware now

  • A licence bill to reset

    Broadcom-era pricing and packaging hit hardest in lab, dev, test and large-tenant environments that use a fraction of the bundle.

  • A network out of segments

    The VLAN standard caps a network at around 4,094 isolated segments, and large multi-tenant estates have used them up.

  • NSX on top of the bill

    Scaled L2 isolation on stock VMware means NSX: more licence cost and more complexity on a bill already in question.

  • Hardware at end of life

    End-of-life switching and storage that no longer fits the workload turn a licence decision into a platform decision.

What we deliver

  • Assessment and proof of concept

    The alternatives weighed against your networking, scale, licensing and operational needs, and the shortlisted platform proved at representative scale before you commit.

  • Platform build as code

    Clusters, storage and automation built as infrastructure-as-code, and plugged into the portals and workflows your teams already use.

  • Workload migration

    VMs moved from ESXi to the target with the network topology kept as it was, after scale and stress testing, so the risk stays low.

  • Network rework

    An EVPN/VXLAN overlay with ~16M segments, switching refreshed on an open network OS such as SONiC, and configuration driven by GitOps.

  • Code-level platform engineering

    Where the chosen hypervisor has gaps in SDN, integration or scale, we close them in source: custom SDN extensions, selective restart, integration daemons.

    Read the case study →
  • Support after the move

    Monitoring extended to the new estate, and ongoing support and managed services for the platform, extensions included.

    Read the case study →

How it runs

The same six stages on every migration. Scope and duration scale with the estate; the stages stay the same.

  1. Assessment

    The alternatives evaluated against your networking, scale, licensing and operational requirements.

  2. Proof of concept

    The shortlisted platform validated at representative scale, with any custom extensions it needs confirmed.

  3. Platform build

    Cluster, storage and automation, fully as code, integrated into your existing portals and workflows.

  4. Migration

    Workloads moved from ESXi to the target platform, with the topology preserved to keep the risk low.

  5. Network rework

    EVPN/VXLAN overlay, physical switching refreshed, and GitOps-driven network automation.

  6. Support

    Managed services for the extended platform, so your team is never left on its own with it.

The target fits your constraints

We weigh five things, in this order: licensing and total cost, SDN and L2 scale, the storage model, proven scale, and operational maturity and APIs.

  • Proxmox VE

    Open source, KVM and LXC with native Ceph. A fit when cost and extensibility lead, and your team is ready to own the stack.

  • Nutanix AHV

    Hyperconverged, with Prism management and bundled support. A fit when operational simplicity matters most.

  • OpenStack

    Modular IaaS with carrier-grade multi-tenancy. A fit for telco, service-provider and large multi-tenant estates.

  • OpenShift Virtualization

    VMs as Kubernetes workloads. A fit when the platform team is Kubernetes-native and wants one control plane.

Technology

  • Platforms we move between

    • VMware ESXi
    • VMware NSX
    • VMware vSAN
    • Proxmox VE
    • Nutanix AHV
    • OpenStack
    • OpenNebula
    • OpenShift
    • KubeVirt
  • Hypervisor and storage

    • KVM
    • LXC
    • Kubernetes
    • Ceph
  • Network and automation

    • EVPN
    • VXLAN
    • FRRouting
    • SONiC
    • Open vSwitch
    • Ansible
    • Terraform
    • Pulumi
    • BGP

Proof

Why CodiLime

  • Code-level platform engineering

    Where the chosen hypervisor has gaps in SDN, integration hooks or scale, we close them in source: per-tenant VLAN pools, selective cluster restart, a netlink-based integration layer.

  • Any target, one programme

    Proxmox, Nutanix, OpenStack, OpenNebula or OpenShift Virtualization: we match the platform to your constraints, and the stages around it stay the same.

  • Virtualization and network in one team

    Infrastructure, network and software engineers and DevOps deliver the hypervisor, the storage and the EVPN/VXLAN fabric together, so the network never waits for another vendor.

FAQ

Good to know

  • Will our users notice the migration?

    The aim is that they do not. The network topology stays as it is while workloads move, the new platform plugs into the portals people already use, and the rollout is tested at scale first. On our largest programme, the end users’ workflow did not change.

  • Do we have to move to open source?

    No. We choose the target to your constraints: Proxmox VE, OpenStack and OpenNebula are open source, Nutanix AHV and OpenShift Virtualization are licensed, and vendor-supported distributions of the open options are on the table too.

  • What if the new platform lacks something we rely on?

    We extend it in source. On one programme we added per-tenant VLAN pools, selective cluster restart and a netlink-based integration layer to stock Proxmox. The proof of concept shows which extensions you need before you commit.

  • Do we have to rebuild the network at the same time?

    Only when the network is the limit. A migration can keep the topology as it is and rework the network in a second phase, which is how our largest programme ran: virtualization first, then EVPN/VXLAN on SONiC.

  • How long does it take?

    It scales with the estate. The assessment and the proof of concept come first, and they end with a plan, a timeline and the extensions you need, before the platform build starts.

  • Who runs the platform afterwards?

    Your team, with the platform built as code and documented, or CodiLime, through ongoing support and managed services for the platform and its extensions.

Start with an assessment

Tell us about your estate. We weigh the alternatives against your networking, scale, licensing and operational needs, prove the shortlisted platform at representative scale, and hand you a plan, a timeline and the extensions you need.