Use case · Web & apps
Kubernetes hosting
Kubernetes is a container orchestration platform that automates deploying, scaling, and managing containerised workloads across a cluster. Hosting it well often means bare metal — avoiding double virtualisation — with self-hosting giving control and European sovereignty over managed alternatives. Kubernetes is powerful but complex, so it is not always the right tool; where it fits, dedicated EU infrastructure suits it.
Key points
- Kubernetes is a container orchestration platform that automates running containers across a cluster.
- It automates deployment, scaling, self-healing, and management of containerised workloads.
- Bare metal suits Kubernetes by avoiding the double virtualisation of running it on cloud VMs.
- Self-hosting gives control and European sovereignty over managed Kubernetes services.
- Kubernetes is powerful but complex, so it is not always the right choice.
What is Kubernetes, and why host it?
Kubernetes is an open-source container orchestration platform that automates deploying, scaling, and managing containerised applications across a cluster of machines. It has become the standard way to run containerised workloads at scale, taking on the work of placing containers on servers, keeping them running, scaling them, and coordinating them, so that applications built from containers can be run and managed in an automated, consistent way. Organisations use Kubernetes to run containerised applications — including microservices and many other workloads — on clusters, relying on it to handle much of the operational work of keeping those containers running well.
Hosting Kubernetes means providing the cluster it runs on and running Kubernetes itself well upon it. Kubernetes runs across a cluster of servers, coordinating containers on them, so it needs infrastructure — servers with the compute, memory, storage, and networking the workloads and Kubernetes require. How that infrastructure is provided, whether bare metal or virtualised, and whether Kubernetes is self-hosted or used as a managed service, shapes the performance, control, and cost of running it. This page focuses on hosting Kubernetes and the choices around it; the architecture of a Kubernetes cluster's components is covered on the cluster page, while what follows attends to why and where to host Kubernetes.
What Kubernetes does
Kubernetes automates much of the work of running containerised applications, which is why it is so widely used for containerised workloads at scale. It places containers onto the cluster's servers according to their needs, restarts them if they fail, scales the number of running instances up or down as load requires, and manages updates to running applications, among much else. This automation means that once workloads are defined for Kubernetes, it handles keeping them running, scaling them, and recovering them from failures, taking on operational work that would otherwise be manual.
This self-healing and automated scaling are among Kubernetes's most valued capabilities. If a container or a server fails, Kubernetes notices and reschedules the affected workloads onto healthy capacity, so the application recovers automatically; and it can scale workloads with demand, running more instances when load rises. For applications that benefit from this automated management — running many containers, needing to scale, or requiring resilience — Kubernetes provides a great deal. The infrastructure beneath Kubernetes must give it the servers and resources across which it does this, which is what hosting Kubernetes provides: the cluster on which Kubernetes automates the running of containerised workloads.
Why bare metal for Kubernetes
Kubernetes runs well on bare metal, and bare metal has a particular advantage for it: it avoids the double virtualisation that results from running Kubernetes on virtual machines. Kubernetes and its containers already provide a layer of abstraction over the hardware; running that on top of virtual machines, as managed cloud Kubernetes often does, adds another virtualisation layer beneath, so the workloads run through two layers rather than one. Running Kubernetes directly on bare metal avoids this doubling, giving the workloads the full performance of the hardware without the overhead of an extra virtualisation layer.
This makes bare metal attractive for Kubernetes where performance and efficiency matter. On bare metal, Kubernetes workloads get the hardware's full performance and the cluster achieves better resource efficiency, since resources are not consumed by an underlying virtualisation layer. For demanding or resource-intensive workloads, and for getting the most from the hardware, running Kubernetes on dedicated bare-metal servers is often preferable to running it on virtual machines. We host Kubernetes on dedicated bare-metal servers, so that the cluster and its workloads run directly on the hardware with full performance and efficiency, avoiding the double virtualisation that running Kubernetes on cloud virtual machines imposes.
Self-hosted versus managed Kubernetes
A significant choice is between self-hosting Kubernetes on infrastructure you control and using a managed Kubernetes service run by a cloud provider. Managed Kubernetes services operate the Kubernetes control plane and much of the cluster for you, which reduces the operational work, but they carry the familiar trade-offs: your workloads and data run on the provider's platform, often under a foreign jurisdiction and on virtual machines; costs depend on the provider; and you are within their ecosystem. Self-hosting Kubernetes on infrastructure you control means operating the cluster yourself but keeps the workloads, data, configuration, and jurisdiction in your hands, and lets you run on bare metal.
The honest basis for choosing is what you value against the operational effort. Managed Kubernetes reduces the work of running the cluster, which is genuinely valuable, and for teams that want that and accept running on the provider's platform, it may suit. Self-hosting requires operating Kubernetes but gives control over the cluster, the performance and efficiency of bare metal, predictable cost, and — importantly — keeping the workloads and data on infrastructure you control under a chosen jurisdiction. Organisations self-host Kubernetes when control, cost, performance, or sovereignty matter enough to justify operating it. We provide the infrastructure for self-hosted Kubernetes, while being candid that managed Kubernetes's reduced operational effort is the right trade for some.
The complexity of Kubernetes
It is worth being honest that Kubernetes is powerful but complex, and it is not always the right tool — for simpler needs, its complexity can outweigh its benefits. Kubernetes brings a substantial learning curve and real operational complexity, since it is a sophisticated system with many concepts and moving parts, and running it well takes expertise. For applications that genuinely need what Kubernetes offers — running many containers, scaling, resilience, and automated management at scale — this complexity is worth it. But for simpler applications that do not need all this, Kubernetes can be more than is warranted, adding complexity for little gain over simpler ways of running containers or applications.
The right choice depends on the workload and the team, and we will not pretend Kubernetes is always the answer. An application that would run fine on a few servers, or in simpler container setups, may not benefit from Kubernetes's power enough to justify its complexity; one running many services at scale, needing automated management and resilience, may benefit greatly. Understanding this helps decide whether Kubernetes fits. We host Kubernetes for those whose workloads warrant it, and we also host simpler container and application setups for those who do not need Kubernetes — matching the approach to the need rather than pushing Kubernetes where it is more than is required.
Scaling and resource efficiency
Kubernetes helps use infrastructure efficiently by packing containers onto the cluster's servers and scaling workloads to their load, which can improve resource utilisation. Because Kubernetes places containers onto servers according to their resource needs, it can use the cluster's capacity well, running many containers across the servers efficiently; and because it scales workloads with demand, it can run more instances when needed and fewer when not, matching resource use to load. On bare metal, this efficiency is greater, since resources are not consumed by an underlying virtualisation layer, so more of the hardware serves the workloads.
This resource efficiency is part of Kubernetes's appeal for running many workloads, since a well-used cluster serves more with given hardware. The efficiency depends on the cluster being sized and used well, which Kubernetes's scheduling and scaling support. For a cluster running substantial containerised workloads, this efficient use of resources contributes to the economics of running them. We provide dedicated bare-metal infrastructure for Kubernetes on which this efficiency is greatest — the hardware's full capacity available to the workloads, without a virtualisation layer consuming resources — so that a Kubernetes cluster hosted with us uses its infrastructure efficiently for the workloads it runs.
Sovereignty, and where VV Internet Hosting fits
Kubernetes runs an organisation's containerised workloads and the data they handle, so where the cluster runs governs that data, making sovereignty a consideration — and self-hosting on EU infrastructure keeps it European. VV Internet Hosting is incorporated in the Netherlands, within the EU, so a Kubernetes cluster hosted with us runs under European jurisdiction and outside the direct reach of the US CLOUD Act, keeping the workloads and their data under European law — unlike managed Kubernetes on a foreign provider's platform. For workloads handling data subject to European protection, self-hosting Kubernetes in the EU keeps that data sovereign.
We host dedicated bare-metal infrastructure for self-hosted Kubernetes: servers with the compute, memory, storage, and networking a cluster and its workloads need, running Kubernetes directly on the hardware for full performance and efficiency, in EU datacenters under European jurisdiction. This suits organisations self-hosting Kubernetes that want the performance and efficiency of bare metal, control over their cluster, and European data sovereignty. We are clear about our limits: we provide the dedicated infrastructure you run Kubernetes on, not a fully managed Kubernetes service that operates the cluster for you; and we are honest that Kubernetes is not always the right tool, hosting simpler setups where it is not needed. We are the dedicated, sovereign infrastructure for self-hosted Kubernetes — and if that fits your workloads, we can host it well.
Questions
Kubernetes, answered plainly
Common questions about hosting for Kubernetes.
What does Kubernetes do?
It automates running containerised applications across a cluster — placing containers on servers, restarting them if they fail, scaling the number of instances with load, and managing updates. This self-healing and automated scaling handle much of the operational work of keeping containers running well, which is why Kubernetes is the standard way to run containerised workloads at scale.
Why run Kubernetes on bare metal?
To avoid double virtualisation. Kubernetes and its containers already abstract over the hardware; running that on virtual machines, as managed cloud Kubernetes often does, adds another virtualisation layer beneath. Running Kubernetes directly on bare metal avoids this doubling, giving workloads the hardware's full performance and better resource efficiency, without an extra layer consuming resources.
Is Kubernetes always the right choice?
No. Kubernetes is powerful but complex, with a real learning curve and operational overhead. For applications that need running many containers, scaling, resilience, and automated management at scale, it's worth it. For simpler applications, its complexity can outweigh its benefits, and simpler ways of running containers or applications may suit better. The right choice depends on the workload and team.
Planning Kubernetes infrastructure?
We host dedicated, EU-sovereign infrastructure sized to your workload — and we will tell you plainly when something else fits better. Tell us what you're building.