Kubernetes platform strategy and architecture
Decide which Kubernetes to run, where to run it, how big it needs to be and who will run it, with a design and a budget you can take to your board.
Get the big decisions right before you buy anything
The most expensive Kubernetes mistakes are made before installation: the wrong distribution for your team's skills, clusters sized for a pilot instead of production, subscriptions bought for cores you don't need, or no plan for who runs the platform once the project team leaves.
Our strategy work answers those questions with evidence. We look at the applications you want to run, your infrastructure, your team and the rules you work under, then recommend a platform, a design, a budget and a roadmap that fit them.
When it fits
01
You are starting with Kubernetes
A first platform is planned, and you want to avoid building something you will have to rebuild in a year.
02
Pilots need to become production
Proof-of-concept clusters work, but they were never designed for security, scale or support.
03
Clusters have multiplied
Several teams or vendors run their own clusters, and you want one platform and one way of working.
04
A renewal or refresh is due
Hardware, subscriptions or a virtualisation renewal is coming up, and the platform choice changes the budget.
What we do
Readiness assessment
Review applications, infrastructure, skills, processes and security against what a production platform needs, and score where you stand.
Platform selection
Compare OpenShift, upstream Kubernetes and managed cloud services on cost, support, skills and data-residency rules, and recommend one.
Reference architecture
Design clusters, sites, networking, storage, identity and security, including how many clusters you need and why.
Sizing and cost
Size nodes, storage and subscriptions from real workload data, and model the cost of each option over three to five years.
Operating model
Decide who runs what between your platform team, your application teams and CompTech, with service levels, on-call and support paths.
Roadmap and business case
A phased plan from pilot to production, with milestones, risks and the benefits that justify the investment.
Readiness assessment
See where your platform stands, and where the roadmap takes it
We score six areas against what a production platform needs. The gap between today and the target becomes the roadmap, ordered to remove the biggest risks first.
- Today
- Where your platform and team stand now, from interviews, configuration reviews and evidence.
- Target
- Where you need to be for your workloads and rules, agreed with you.
- Six measures
- Skills, automation, security, observability, resilience and governance.
How a strategy assessment runs
01
Discover
Workshops with your application, infrastructure, security and operations teams, and a review of what you run today.
02
Analyse
Score readiness, profile the workloads, and compare the platform options on cost, risk and fit.
03
Design
Draw the target architecture and operating model, and size the hardware, subscriptions and team.
04
Plan
Agree a phased roadmap, budget and business case, ready for approval and procurement.
What you get
- A readiness score across six areas, with the evidence behind it
- A platform recommendation, with the options compared
- A reference architecture for clusters, sites, networking, storage and security
- Sizing for servers, storage and subscriptions
- Cost models for each option over three to five years
- An operating model with roles and service levels
- A phased roadmap and business case
What we compare
Each option is weighed on your workloads, skills, support needs and budget.
Platforms
- Red Hat OpenShift
- Upstream Kubernetes
- Amazon EKS
- Azure Kubernetes Service
- Google Kubernetes Engine
- OpenShift on AWS and Azure
Where it runs
- Bare-metal servers
- VMware vSphere
- OpenStack
- Public cloud
- Branch and edge sites
Questions about platform strategy
How long does a strategy assessment take?
Weeks, not months. The length depends on how many applications, sites and teams are in scope, and we agree it with you at the start.
Will you just recommend OpenShift because you sell it?
We are a Red Hat partner, and we also build upstream Kubernetes and managed cloud platforms. The recommendation follows the evidence on your workloads, team, support needs and budget, and we show you the comparison so you can check it.
Can we use hardware we already own?
Often, yes. We check whether your servers, storage and networks meet the platform's needs and reuse what fits. The sizing shows what, if anything, you need to buy.
What happens after the assessment?
You own the design and the roadmap, and can take them to any supplier. Many clients ask us to build the platform, which we run as a separate project.
Platform implementationPlan your platform with evidence
Tell us what you want to run and the constraints around it. We'll propose the scope of an assessment and what you will have at the end of it.