01 / VKS migration services
If you run VCF, you already own a full Kubernetes platform: vSphere Kubernetes Service (VKS). We move your Kubernetes workloads onto it. We prove it on one of your own workloads first, then migrate the rest in reversible waves.
Broadcom Expert Advantage
Partner
4–6 weeks
To one workload live
Rollback
At every gate
02 / Where you start
Most start from one of these situations. Each has its own plan, and its own business case.
03 / Why act now
Every figure below comes from a named, dated survey, so you can check it.
82%
of container users run Kubernetes in production, up from 66% in 2023.
CNCF Annual Cloud Native Survey, Jan 2026 · source
88%
reported a year-on-year rise in their total Kubernetes cost of ownership.
Spectro Cloud, State of Production Kubernetes 2025 (455 respondents) · source
29%
of cloud spend is estimated to be wasted, the first increase in five years.
Flexera 2026 State of the Cloud Report, Mar 2026 · source
50%
of enterprises have already moved some workloads from public to private cloud, up from 35% a year earlier.
Broadcom Private Cloud Outlook 2026 (1,800 IT decision-makers) · source
04 / Why VKS on VCF
VKS is part of VMware Cloud Foundation, so there is no second Kubernetes subscription to buy or renew.
Kubernetes runs on the vSphere, storage and networking you already operate, managed by the team you already have.
CNCF-conformant Kubernetes, so standard workloads move as they are and stay portable.
Registry, GitOps, backup and policy services come packaged with VKS, with Avi, vDefend and Data Services Manager available as add-ons.
Advanced servicesAdd-ons; may be licensed separately
VKS add-onsPackaged with VKS
vSphere Kubernetes ServiceCNCF-conformant, on vSphere Supervisor
VMware Cloud FoundationWhat you already run
05 / How we work
Each step earns the next. VKS runs alongside your current platform until you decide to move.
What VKS in your VCF already covers, compared with what you run today.
30 minutes · VCF and platform leads
Your platforms, candidate workloads and success criteria. Ends with an agreed pilot scope and sponsor.
About 3 hours · infra, platform and app owners
VKS activated on your VCF and one workload live, measured against the criteria you set.
4–6 weeks · fixed scope
Inside the pilotThe rest of your workloads move on your schedule, with rollback at every gate. The old platform is retired.
Planned from each workload's readiness
First waves
Portable
Moves as it is: standard manifests and Helm charts.
Middle waves
Patchable
Small changes: ingress, storage classes, registry paths or policies.
Final waves
Convertible
Needs rework: platform-specific APIs, operators or stateful data.
Every workload is scored during the assessment, so the number of waves and the timeline reflect your applications, not a generic package. The pilot may be eligible for funding through your Broadcom services entitlement; ask us.
06 / Inside the pilot
A fixed-scope pilot that activates VKS on the VCF you already run, moves one real workload onto it and measures the result against criteria you set.
You get: every workload scored portable, patchable or convertible.
You get: a signed-off pilot design and measurable criteria.
You get: one workload live and a results readout.
07 / Typical timeline
A typical plan. We adjust the weeks to your environment and change windows.
Week 0 is the discovery workshop. The pilot itself runs weeks 1–6; a reference-app pilot can finish in about four.
08 / Who is involved
09 / After the pilot
The pilot is the Crawl stage. The migration that follows uses the same eight-step, reversible method, wave by wave.
Consolidation
Stand up VKS next to the current platform, inventory workloads and dependencies, migrate one pilot workload.
Repatriation
Build a VKS landing zone on spare VCF capacity, capture the cloud cost baseline, move one steady-state workload.
Done when: one workload runs on VKS and is measured against the baseline.
Consolidation
Move stateless workloads in waves, lowest risk first, and repoint GitOps and pipelines to VKS.
Repatriation
Move always-on stateless workloads first, running hybrid with the cloud clusters.
Done when: waves run on a schedule and GitOps targets VKS.
Consolidation
Move the remaining stateful workloads, then decommission the old platform and let its subscription lapse.
Repatriation
Move data-bearing workloads, using Data Services Manager for databases, and shrink cloud clusters before the commitment renews.
Done when: VKS is the governed default for new and steady-state workloads.
10 / What you get
Running on the VCF you already own, measured against your own success criteria.
Every workload scored portable, patchable or convertible, with Wave 1 chosen.
Savings, migration effort and payback built from your numbers, not a generic model.
Sponsor, infrastructure, platform and application owners agree on the results.
Knowledge transfer, Day-2 runbooks and a self-service namespace process.
11 / Why AxelCore
Moving to VKS is part platform work and part application work. The same people who switch on VKS also move your applications, repoint your pipelines and hand over a platform your teams can run.
Every engagement ends with a handover: hands-on knowledge transfer, Day-2 runbooks and a self-service namespace process, so your team runs VKS without us.
12 / The business case
We build your business case with you from a few numbers gathered in discovery, and show which of them decide the result.
Savings start at the renewal date of the subscription you no longer need.
Savings start with the always-on clusters, whose cloud cost is fixed and visible.
13 / Questions
VKS is part of VMware Cloud Foundation. Advanced services such as Avi Load Balancer, vDefend and Data Services Manager may be licensed separately. In the capability review we confirm what your own licence covers.
No single network design is forced on you. VKS can run with NSX or with vSphere networking and a supported load balancer, and we design around the network you already run, including Cisco ACI environments.
Rancher, OpenShift, TKGi, EKS, AKS, GKE and self-built Kubernetes. Each workload is classified as portable, patchable or convertible before anything moves.
Yes. Repatriation targets the always-on baseline. Burst workloads and those tied to cloud-native services can stay where they are.
That is the goal of every engagement. We work alongside your engineers and end with knowledge transfer, Day-2 runbooks and a self-service process for new namespaces.
Get started
Start with a 30-minute capability review. We show you what your VCF licence already covers and whether a pilot makes sense for you.