Teaching Cluster Autoscaler to speak cloudscale.ch without forking the autoscaler or waiting for an upstream provider integration.

Introduction

What do you do when Kubernetes Cluster Autoscaler supports the scaling logic you need, but not the cloud provider you run on?

That was the problem behind this experiment. I wanted Cluster Autoscaler to work on cloudscale.ch. There was no upstream provider for it, and I did not want to fork the autoscaler or carry an in-tree provider forever.

The answer was externalgrpc: a cloud provider integration mode that lets Cluster Autoscaler keep its core decision logic while delegating cloud-specific actions to a separate gRPC service.

This post walks through how Cluster Autoscaler makes scaling decisions, what the provider contract looks like, how externalgrpc fits into the architecture, and the practical gotchas I ran into while building an autoscaler provider for cloudscale.ch.

The autoscaler knows when, not how

The first important mental model is this:

Cluster Autoscaler knows when the cluster should scale, but it does not inherently know how to create or delete machines.

The core autoscaler logic decides things like:

• Are there pending Pods that cannot be scheduled anywhere?
• Would those Pods fit on a node from a configured node group?
• Has a node been idle long enough that its Pods could be moved elsewhere?
• Which node group should be increased or decreased?

What it does not know is how to create a VM, which image to use, how to pass bootstrap data, how the node joins the cluster, or how to delete the machine again.

That part is delegated to a cloud provider plugin.

For large providers, there are already in-tree implementations. But if your provider is not on the list, you need another strategy.

What happens every 10 seconds

By default, Cluster Autoscaler runs a reconciliation loop roughly every 10 seconds. In each loop, it asks a set of questions:

1. Are there unschedulable Pods?
2. Would those Pods fit on a node from a configured node group?
3. If yes, call NodeGroupIncreaseSize(group, delta).
4. Are there nodes that have been idle for long enough?
5. Can their Pods be rescheduled elsewhere?
6. If yes, call NodeGroupDeleteNodes(group, nodes).

The provider does not make these decisions. The provider only receives the result of the decision, for example: “increase this node group by three nodes.”

That separation is useful. It means a provider implementation can stay relatively small. The hard Kubernetes scheduling logic remains in Cluster Autoscaler.

The simulator is the scheduler

Cluster Autoscaler does not reimplement Kubernetes scheduling from scratch. It imports the Kubernetes scheduler framework and uses the same kinds of filter logic the scheduler uses.

That includes scheduling constraints such as:

• node resource fit
• node affinity
• taints and tolerations
• topology spread constraints
• inter-pod affinity
• volume binding
• node ports
• volume limits

The autoscaler runs these checks against an in-memory snapshot of the cluster. For scale-down, it can simulate removing candidate nodes and see whether their Pods would fit elsewhere. For scale-up, it can simulate whether pending Pods would fit on a hypothetical node from a node group.

That is where one of the most important provider methods comes in: NodeGroupTemplateNodeInfo.

Node groups, flavours, and choosing what to scale

If the cloud provider supports multiple VM types, sizes, or flavours, the provider usually exposes those as separate node groups. For example, one node group might represent small general-purpose workers, another might represent larger memory-optimized workers, and another might represent nodes with different labels or taints.

Cluster Autoscaler then runs the simulation against the configured node groups. If more than one group could fit the pending Pods, the configured expander decides which group wins. Depending on configuration, that can be random, most-pods, least-waste, price, priority, or another supported strategy.

This gives you a useful amount of flexibility, but it is still not Karpenter. Cluster Autoscaler chooses from node groups that already exist as configured abstractions. Karpenter has a more dynamic provisioning model and can reason more directly about instance types, constraints, and provisioning choices.

So the mental model is:

• Cluster Autoscaler: choose between predefined node groups
• Karpenter: more dynamic node provisioning based on workload requirements and available instance types

For an externalgrpc provider, this means the provider should model each relevant cloudscale.ch flavour as a node group, return accurate template node information for each one, and let the autoscaler’s simulator and expander decide which group should grow.

Why externalgrpc?

If your cloud provider is not supported upstream, there are three realistic options:

Approach	What it means	Trade-off
Upstream in-tree provider	Add your provider to kubernetes/autoscaler	High review and maintenance burden, tied to upstream release cycles
Fork Cluster Autoscaler	Maintain your own autoscaler build	Maximum control, but you own the fork forever
externalgrpc	Run a standalone provider service over gRPC	Separate release cycle, language-agnostic, small provider surface

For cloudscale.ch, externalgrpc was the pragmatic option.

It avoids the need to become an upstream autoscaler maintainer. It avoids carrying a fork. And because the provider is just a gRPC service, it can be implemented in any language. In this case, I used Go.

At the time of writing, it also aligns well with where the autoscaler project appears to be heading. The ongoing SIG Autoscaling refactor proposal points toward splitting provider implementations out of the main autoscaler repository. In that future, providers would either consume the autoscaler core as a library or communicate out of process via externalgrpc.

For an independent provider implementation, that makes externalgrpc a good escape hatch.

Architecture

At a high level, the architecture looks like this:

Pending Pods
    |
    v
Cluster Autoscaler
    |
    | gRPC over mTLS
    v
autoscaler-cloudscale
    |
    | cloudscale.ch REST API
    v
cloudscale.ch VMs
    |
    v
VM boots, kubelet joins, CCM stamps providerID

The Cluster Autoscaler remains the brain. It watches the Kubernetes cluster, notices pending Pods, runs scheduling simulations, and decides whether a node group should grow or shrink.

The external provider is the hands. It receives requests from Cluster Autoscaler and performs cloud-specific actions:

• list cloudscale.ch servers
• create VMs
• tag VMs
• delete VMs
• map Kubernetes nodes back to cloudscale.ch servers

The provider also needs configuration for node groups, for example:

nodeGroups:
  - name: worker
    flavor: flex-8-2
    min: 1
    max: 5
    image: custom:talos-1.13.2
    userData: @talos
    tag: k8s-autoscaler-group

In my setup, the VM bootstrap is handled through Talos Linux machine configuration passed as user data. The VM boots, Talos configures the node, kubelet joins the cluster, and the cloud controller manager stamps the node with a provider ID.

That provider ID is critical. Without it, the autoscaler cannot reliably connect a Kubernetes Node back to the corresponding cloud VM. Scale-down depends on this mapping.

The provider contract: six important RPCs

The externalgrpc proto contains more methods, but the core provider behavior is built around a small set of RPCs:

RPC	Purpose
Refresh	Reconcile provider state with the cloud
NodeGroups	Return configured node groups
NodeGroupForNode	Map a Kubernetes node to a node group
NodeGroupTemplateNodeInfo	Describe what a new node would look like
NodeGroupIncreaseSize	Create new servers for a node group
NodeGroupDeleteNodes	Delete specific servers

The remaining methods are either bookkeeping or can be implemented as explicit Unimplemented stubs, as long as the service satisfies the proto contract.

The most important design decision is to make the cloud provider the source of truth. Do not trust only local state. VMs can be manually deleted, failed creates can leave partial state, and API calls can fail halfway through.

In practice, Refresh lists servers from the cloud provider API, filtered by tags, and rebuilds the provider’s internal view.

Conceptually:

servers, err := api.Servers.List(ctx, cloudscale.WithTagFilter(tags))
if err != nil {
    return err
}

byUUID := make(map[string]*cloudscale.Server, len(servers))
for i := range servers {
    byUUID[servers[i].UUID] = &servers[i]
}

cache.Store(byUUID)

Tags are what scope the provider to the right cluster and node group. In this implementation, group membership is represented with a tag such as:

k8s-autoscaler-group=<name>

TemplateNodeInfo: modelling future nodes

When scaling from zero, there may be no real worker node in the cluster. So how can Cluster Autoscaler know whether a pending Pod would fit on a node that does not exist yet?

The provider has to describe what a future node would look like.

NodeGroupTemplateNodeInfo returns a synthetic Kubernetes Node object. This object tells the autoscaler things like:

• CPU capacity
• memory capacity
• disk-related properties
• labels
• taints
• allocatable resources
• maximum Pod count
• special devices or constraints

This is where the provider has to “lie” to the autoscaler, but only in a very specific sense. The node is not real yet. It is a placeholder used for scheduling simulation.

As an analogy, this is conceptually similar to the idea of virtual nodes: an object that represents capacity that is not currently a normal worker node. This is only an analogy, not Virtual Kubelet and not a real virtual-node implementation. Cluster Autoscaler does not schedule Pods onto this synthetic node. It uses the template to answer the question: “If I created a node like this, would these pending Pods fit?”

In other words, the template node is not runtime capacity. It is simulated future capacity.

One easy mistake is to expose the raw VM flavor as the node capacity without accounting for kubelet reservations, system reservations, or eviction thresholds. A VM with 8 GiB of memory does not have 8 GiB of allocatable memory for workload Pods.

Another easy mistake is forgetting to set ResourcePods. If the template node says it can run zero Pods, the autoscaler will correctly conclude that no workload can ever fit there.

That bug is wonderfully frustrating: everything looks wired up, the autoscaler runs, but nothing scales.

Creating nodes: target size first, VM later

When Cluster Autoscaler decides to increase a node group, it calls NodeGroupIncreaseSize with a delta.

The provider then needs to:

1. Increase the target size in its own node group state.
2. Create the requested number of VMs.
3. Tag them correctly.
4. Pass the bootstrap configuration.
5. Wait for the nodes to appear in Kubernetes.
6. Let Refresh reconcile the actual cloud state again.

It is important to distinguish three different states: desired capacity, in-flight capacity, and joined Kubernetes nodes. A VM may be requested but not yet visible, booting but not yet joined, or joined but not yet Ready. During that period, Cluster Autoscaler must not repeatedly create more and more machines every 10 seconds.

Cluster Autoscaler also has guardrails for failed provisioning. If a node does not join within the configured provisioning duration, the autoscaler can treat that scale-up as failed and move on.

Bootstrapping with Talos

In this implementation, node bootstrap is intentionally kept outside the autoscaler’s core logic.

The node group config contains the Talos machine configuration as user data. When autoscaler-cloudscale creates a VM, it passes that user data to cloudscale.ch. Talos then handles the rest:

• boot the custom image
• apply machine configuration
• use the configured certificates and join tokens
• start kubelet
• join the Kubernetes cluster

This keeps the provider focused on VM lifecycle management rather than becoming a full cluster lifecycle engine.

The one hard requirement is that the cluster has a working cloud controller manager. It must stamp Kubernetes nodes with a provider ID, because that is the bridge between Kubernetes objects and cloudscale.ch servers.

No provider ID means no reliable scale-down.

Demo: scale from zero workers

The demo shows a Kubernetes cluster on cloudscale.ch starting with only a control plane node and no workers. Workload pressure is then applied to trigger the autoscaler and the external provider.

To generate that pressure, the demo deploys 50 replicas of the Kubernetes pause container with small CPU and memory requests – a deliberately boring Deployment that exists only to create scheduling demand.

apiVersion: apps/v1
kind: Deployment
metadata:
  name: scale-demo
spec:
  replicas: 50
  selector:
    matchLabels:
      app: scale-demo
  template:
    metadata:
      labels:
        app: scale-demo
    spec:
      containers:
        - name: pause
          image: registry.k8s.io/pause:3.9
          resources:
            requests:
              cpu: 500m
              memory: 128Mi

The Pods could not be scheduled because there were no worker nodes. Cluster Autoscaler noticed the pending Pods, simulated whether they would fit on a node from the configured worker group, and called the external provider.

The provider created three cloudscale.ch VMs. They booted the Talos image, installed to disk, rebooted, joined the cluster, and eventually became Ready. At that point, Kubernetes could schedule some of the pending Pods onto the new workers.

The important part of the demo is not the exact terminal output. It is the separation of responsibilities:

• Kubernetes produced pending Pods.
• Cluster Autoscaler simulated future capacity using TemplateNodeInfo.
• The expander selected a configured node group.
• The externalgrpc provider created the cloudscale.ch VMs.
• Talos bootstrapped the nodes into the cluster.
• The CCM stamped provider IDs so the autoscaler could map Kubernetes nodes back to cloud VMs.

That makes the demo a useful validation of the whole architecture without turning the blogpost into a screen-recording transcript.

Gotchas

1. TemplateNodeInfo is easy to get subtly wrong

The autoscaler’s decision quality depends heavily on the synthetic node returned by NodeGroupTemplateNodeInfo.

If allocatable CPU or memory is too optimistic, the autoscaler may under-scale. If it is too pessimistic, it may over-scale. If ResourcePods is missing or zero, nothing will schedule at all.

This method feeds both scale-from-zero and the internal “upcoming node” logic. Treat it as part of the scheduling model, not as a cosmetic metadata method.

2. Tags are not optional

Tags are how the provider knows which cloud VMs belong to which cluster and node group.

Without consistent tags, Refresh cannot reliably reconcile state. Scale-down becomes dangerous because the provider cannot know which machine corresponds to which Kubernetes node.

3. The cloud is the source of truth

Local state is useful, but it is not authoritative.

Manual deletes, failed creates, partial outages, quota issues, and API timeouts all happen. Refresh should reconcile against the provider API every loop and rebuild the cache from reality.

4. Lock ordering matters

Provider implementations often keep shared caches plus per-node-group state. If multiple mutexes are involved, lock ordering needs to be consistent.

One subtle bug here was a potential deadlock between Refresh and target-size updates. The fix was to always acquire locks in the same order.

5. No CCM, no reliable scale-down

A cloud controller manager is not just a nice addition. It is what stamps the Kubernetes node with the providerID that lets the autoscaler map a Kubernetes Node to a cloud VM.

Scale-up can appear to work without a complete mapping. Scale-down will not be safe.

Takeaways

Writing a Cluster Autoscaler provider is less about reimplementing autoscaling and more about providing the missing cloud-specific glue.

Cluster Autoscaler already knows how to:

• detect pending Pods
• simulate scheduling
• account for DaemonSets
• apply backoff
• choose node groups
• avoid duplicate scale-ups
• decide when nodes can be removed

The provider needs to know how to:

• describe a future node accurately
• create VMs
• delete VMs
• tag resources
• reconcile cloud state
• map Kubernetes nodes back to cloud resources
• bootstrap machines into the cluster

With externalgrpc, that provider can live as a standalone binary with its own release cycle. For unsupported clouds, lab environments, private infrastructure providers, or sovereign cloud platforms, that is a very practical integration model.

In this case, the result was a small Go service that taught Cluster Autoscaler how to operate on cloudscale.ch without modifying the autoscaler itself.

The autoscaler remained the brain. The external provider became the hands.

That is the real value of externalgrpc. It is not the most glamorous integration point, but it is exactly the right abstraction when you want to keep autoscaling logic upstream and put cloud-specific lifecycle logic in your own hands.

If you’d like to dive deeper, check out the blog post on my personal blog, where I explore the topic in more detail.

Kubernetes is a powerful platform for deploying and managing containerized applications at scale, and it has become increasingly popular in Switzerland in recent years.

One way to approach it is outsourcing. This can be a strategic and cost-effective option for organizations that do not have the in-house DevOps expertise, know-how, or resources to manage their infrastructure and application operations efficiently.

Not every tech company is in the business of building platforms and operating Kubernetes clusters. Thus by partnering with an experienced partner, companies can tap into a wealth of knowledge and expertise to help them run their applications.

Some companies adopt Kubernetes and look to leverage its capabilities themselves. It’s essential to consider time, effort, and possible implications while utilizing the latest developments and continually adding value to the core business.

In all cases, it will be helpful to align with fundamentals. For this reason, I have compiled a quick guide to Kubernetes in 2023 and best practices in Switzerland.

1. Understand the basics: Before diving into Kubernetes, have a solid understanding of the reasoning and concepts. This could include cloud infrastructure, networking, containers, how they liaise with each other, and how they can be managed with Kubernetes.
2. Plan your deployment carefully: When deploying applications with Kubernetes, you must plan thoroughly and consider your workloads‘ specific needs and requirements. This includes but is not limited to resource requirements, network connectivity, scalability, latency, and security considerations.
3. Use appropriate resource limits: One of the critical benefits of Kubernetes is its ability to manage resources dynamically based on the needs of your applications. To take advantage of this, try to set appropriate resource limits for your application. This will help ensure that your application has the resources they need to run effectively while preventing them from consuming too many resources and impacting other applications.
4. Monitor your application: It’s essential to monitor your applications and the underlying Kubernetes cluster to ensure they are running smoothly and to identify any issues that may arise. You want to analyze the alerts and react accordingly. You can use several tools and practices to monitor your applications, including log analysis, monitoring with tools like Prometheus and Grafana, and alerting systems.
5. Use appropriate networking configurations: Networking is critical to any Kubernetes deployment, and choosing the proper network configuration is substantial. What about load balancing, service discovery, and network segmentation?
6. Secure your application: Security is a top concern for many companies and organizations in Switzerland. You cannot proceed without ensuring that your Kubernetes deployment is secure. At this stage, your team is implementing network segmentation, using secure container runtime environments, and implementing advanced authentication and authorization systems.
7. Consider using a managed Kubernetes service: For companies without the resources or needing DevOps expertise to manage their clusters, managed Kubernetes services can be a business-saving solution. With managed services, you can get a production-ready cluster, i.e., a fully-managed Kubernetes environment, allowing teams and software engineers to focus on developing new features and deploying their applications rather than managing the underlying infrastructure.
8. Stay up-to-date with the latest developments: The Kubernetes ecosystem is constantly evolving, and it’s better to stay up-to-date with the latest developments and best practices. This may involve subscribing to newsletters like VSHN, VSHN.timer, or Digests, attending conferences and CNCF meetups, and following key players in the Kubernetes community.

By following best practices, IT leaders, stakeholders, and decision-makers can ensure that they use Kubernetes constructively and get the most out of the platform technology.

Update: hier findest du unsere Kubernetes Distributionen 2026 Übersicht.

Bei VSHN und APPUiO.ch setzen wir auf OpenShift als Kubernetes Distribution. Was eine Kubernetes Distribution ist, warum wir das machen und wo die Unterschiede zu „plain“ Kubernetes liegen erklären wir in diesem Blogpost.

Was ist Kubernetes?

Die offizielle Beschreibung von Kubernetes lautet:

Kubernetes is a portable, extensible open-source platform for managing containerized workloads and services, that facilitates both declarative configuration and automation.

Der wichtigste Teil daraus ist die Tatsache, dass es sich um eine Plattform handelt und nicht um ein fertiges Produkt ab der Stange. Für das Verständnis dieses Beitrags ist dies eine wichtige Information.

Was ist eine Kubernetes Distribution?

Um die Unterschiede von Kubernetes und OpenShift zu verstehen, gilt es zuerst einmal den Begriff „Kubernetes Distribution“ zu klären: Wird Kubernetes direkt aus dem Open Source Kubernetes Projekt installiert, erhält man „nur“ die Kernkomponenten (API Server, Controller Manager, Scheduler, Kubelet, kube-proxy). Damit Kubernetes aber auch wirklich benutzbar wird, benötigt man sehr viele weitere Komponenten wie etcd, Ingress Controller, Logging Server, Metrics Collector (z.B. Prometheus), Software Defined Network (SDN) und viele mehr. Dies ist sehr gut vergleichbar mit Linux: Der Linux Kernel alleine bringt noch nicht viel, es braucht eine ganze Linux Distribution welche eine Shell, das Paketmanagement, den Bootprozess und vieles mehr zur Verfügung stellt.

OpenShift ist eine Kubernetes Distribution und macht aus Kubernetes ein Produkt

Eine „Minimum Viable Kubernetes Distribution“ benötigt folgende zusätzlichen Komponenten und Tools für einen produktiven Betrieb:

• Installations- und Upgrademechanismus: Für eine automatisierte Installation aller involvierten Komponenten.
• SDN (Software Defined Network): Pods müssen untereinander kommunizieren können, egal wo sie laufen. Dies stellt das SDN sicher.
• Ingress Controller: Damit Benutzerzugriff auf die auf dem Cluster laufende Applikationen möglich ist.
• Authentication: Eine zentrale Benutzer- und Gruppendatenbank stellt den authentisierten und autorisierten Zugriff zur Verfügung.
• Security: Kubernetes führt Container via Docker oder CRI-O aus. Die Sicherheit auf dem Hostsystem muss entsprechend gewährleistet sein.
• Persistent Storage: Stateful Applikationen wie Datenbanken benötigen persistenten Storage.
• Monitoring: Ständige Überwachung aller Clusterkomponenten und Applikationen.
• Backup: Sicherung der Clusterkomponenten und Persistenten Daten.

Optional werden weitere Komponenten empfohlen:

• Zentrales Logging mit grafischer Aufbereitung und Durchsuchbarkeit
• Applikations- und Cluster Metrics inkl. Alerting

OpenShift als Kubernetes Distribution

Im Kern basiert OpenShift zu 100% auf Kubernetes, bringt aber als Kubernetes Distribution alles mit, was zur Benutzung eines Kubernetes Clusters benötigt wird. Um nur die wichtigsten Funktionen zu nennen:

• Operations Tools: Ein offizieller und unterstützter Weg via Ansible ermöglicht den gesamten Lifecycle von OpenShift durchzuführen. Dazu gehört die automatisierte Installation, wie auch Upgrades auf neuere Versionen von OpenShift.
• Router: Der OpenShift Router (Ingress Controller) – basierend auf HAProxy – sorgt dafür, dass der Zugriff auf Applikationen innerhalb des Clusters über HTTP(S) ermöglicht wird.
• Multi-Tenancy: Die im Kern eingebaute Multi-Tenancy über OpenShift Projekte, RBAC und weitere Konzepte ermöglicht die Benutzung der Plattform durch verschiedene Stakeholder.
• Authentication: Es werden die unterschiedlichsten Authentication Backends unterstützt, allen voran LDAP, ActiveDirectory und weitere.
• Metrics: Die mitgelieferte Metrics Komponente sammelt alle verfügbaren Messwerte (RAM, CPU, Netzwerk) der auf dem Cluster laufenden Applikationen und visualisiert diese in der Webkonsole.
• Central Logging: Alle von der Applikation auf stdout geloggten Zeilen werden automatisch von der zentralen Logging Komponente gesammelt und über die Webkonsole dem Benutzer zur Verfügung gestellt.
• Security: Die Plattform ist auf höchste Sicherheit ausgelegt. So sorgen z.B. Sicherheitsmassnahmen im Kernel von Red Hat Enterprise Linux wie SELinux dafür, dass die Sicherheit der Container gewährleistet ist. Weitere Massnahmen wie „Security Context Constraints“ (SCC) und das Verhindern von Root Containern sorgen für weitere Sicherheit.
• Builds und Pipelines: Direkt im Cluster integrierte Build- und Pipeline-Funktionalitäten ermöglichen einen komplett integrierten CI/CD Workflow.
• Webkonsole: Alle Vorgänge auf dem Cluster werden für den Anwender der Plattform in einer Webkonsole graphisch dargestellt und ermöglichen einen einfachen und schnellen Einstieg in die Benutzung von Kubernetes.
• SDN: Das mitgelieferte Software Defined Networking sorgt für die Konnektivität zwischen den auf der Plattform laufenden Pods und für eine angemessene Netzwerksicherheit mit Network Policies.
• Container Registry: Docker / Container Images werden in der mitgelieferten Registry gespeichert und zum Deployment auf die Worker Nodes benutzt.

Alle diese von Haus aus mitgelieferten Funktionalitäten lassen sich zu jedem Kubernetes Cluster hinzufügen, was jedoch mit einem hohem Aufwand verbunden ist. Vergleichbar mit dem Bau einer eigenen Linux Distribution, wie das z.B. Linux From Scratch veranschaulicht. Für Kubernetes existiert eine ähnliche Anleitung, genannt Kubernetes The Hard Way.

OpenShift als PaaS

Die Stärke von Kubernetes liegt in der Container Orchestrierung. Zusätzlich dazu bietet OpenShift klassische Platform-as-a-Service (PaaS) Funktionen. Eine davon ist das automatische builden und deployen von Applikationscode direkt ab einem Git Repository. Trotzdem hat man als Anwender der Plattform dank der grossen Flexibilität aber immer die Wahl, ob man die integrierten Buildfunktionen nutzen möchte, oder doch lieber ausserhalb des Cluster builden möchte. Dies lässt sich für jedes Deployment entscheiden, so können auf einem Cluster beide Arten verwendet werden.

OpenShift als Upstream zu Kubernetes

Viele Entwicklungen in Kubernetes stammen ursprünglich aus OpenShift. Als bestes Beispiel lässt sich RBAC (Role Based Access Control) nennen. Dieses Feature ist seit der ersten OpenShift-Version Bestandteil und wurde sukzessive in Kubernetes eingebaut. RBAC ist seit Kubernetes Version 1.6 fester Bestandteil von Kubernetes. Auch das OpenShift „Route“- oder das „DeploymentConfiguration“-Objekt hat die heutigen Objekte „Ingress“ bzw. „Deployment“ in Kubernetes massgeblich mitgeprägt.
Da OpenShift zu 100% auf Kubernetes basiert, werden auch alle Kubernetes Native Workloads unterstützt, wie z.B. das „Deployment“- oder das „Ingress“-Objekt.
Schaut man etwas genauer auf die Contributor-Statistiken stellt man fest, dass Red Hat zu den Top 3 Contributor-Firmen gehört, somit ist Red Hat massgeblich an der Entwicklung von Kubernetes beteiligt. Mit dem Kauf der Firma CoreOS hat sich Red Hat geballtes Kubernetes Know-how angeeignet. Die Verschmelzung von OpenShift und Tectonic wird der nächste Meilenstein der Kubernetes Distribution OpenShift sein.

Alternativen zu OpenShift

OpenShift ist nicht die einzige Kubernetes Distribution auf dem Markt. Ein kurzer Vergleich zeigt die Unterschiede:

• Cloud Vendor Kubernetes: Die grossen Clouds bieten ihre eigenen Kubernetes Distributionen als Service an. Diese sind auf die jeweiligen Clouds zugeschnitten und werden von den Anbietern gepflegt. Eine Installation auf der eigenen Private Cloud oder auf anderen Public Clouds ist nicht möglich.
  • Amazon Elastic Container Service for Kubernetes (Amazon EKS)
  • Google Kubernetes Engine
  • Azure Kubernetes Service (AKS)
  • IBM Cloud Kubernetes Service
  • Alibaba Cloud Container Service for Kubernetes
• Rancher: Seit der Version 2.0 fokussiert sich Rancher zu 100% auf Kubernetes und bietet als grosse Stärke eine Multi-Cluster Verwaltungsfunktion. So können mit Rancher Kubernetes Cluster in der Cloud (z.B. auf Amazon oder Google) zentral verwaltet werden, wie auch Kubernetes Cluster mit der „Rancher Kubernetes Engine“ auf eigenen VMs. Mit dem Webinterface gestaltet sich das aufsetzen eines neuen Clusters sehr einfach und Applikationsdeployments mittels Helm sind auch direkt verfügbar.
• Tectonic: Diese Distribution legt grossen Wert auf Cloud-Native-Automatisierung. Durch den Kauf von CoreOS durch Red Hat wird Tectonic mit OpenShift zusammengeführt und viele der Funktionen werden von Tectonic in OpenShift eingebracht.
• Canonical / Ubuntu Kubernetes: Plattform basierend auf Ubuntu, welches Juju als Installationstool verwendet. In Partnerschaft mit Google und Rancher wird in Zukunft eine Hybrid-Cloud-Lösung angeboten.
• SUSE CaaS-Plattform: Eine sehr neue Plattform, basierend auf SUSE MicroOS. Mittels Salt wird die Konfigurationsverwaltung sichergestellt. Unter folgendem Link kann am Beta Programm teilgenommen werden: SUSE CaaS Platform Beta.

Weitere Kubernetes Distributionen aufgelistet:

• Kontena Pharos
• Cisco Container Platform
• Und noch mehr aus der Kubernetes Dokumentation: Picking the Right Solution

Ein sehr wichtiger zu beachtender Aspekt ist der Cloud- und/oder Vendor-Lock-In. Viele der Kubernetes Distributionen haben ihre eigenen Eigenschaften, die unter Umständen nicht miteinander kompatibel sind. Am Beispiel der „Cloud-Vendor“-Distributionen: Diese können nur in der entsprechenden Cloud benutzt werden. Möchte man jedoch einen Hybrid-Cloud-Ansatz verfolgen, ist dies durch den Lock-In nicht möglich. Im Gegenzug ermöglicht eine selber installierbare Distribution wie OpenShift diese Option.
Reine Open Source Distributionen ohne Herstellersupport sind für produktive Umgebungen nicht zu empfehlen, da dieser für eine komplexe Plattform wie Kubernetes von grossem Vorteil ist.

APPUiO – Swiss Container Platform

Dem aufmerksamen Leser ist bestimmt aufgefallen, dass zwischen der „Minimum Viable Kubernetes Distribution“ und OpenShift gewisse Diskrepanzen bestehen. Genau dort setzt APPUiO an: Wir veredeln OpenShift zu einer vollständigen, production-ready Kubernetes Distribution indem wir Managed Services anbieten. Wir überwachen und sichern den Clusterstatus automatisch, kümmern uns um regelmässige Updates, beheben Fehler, stellen Persistent Storage zur Verfügung und helfen mit unserem Know-how das Beste aus der Plattform herauszuholen.

Weitere Informationen zum Thema Kubernetes und OpenShift

Am Cloud Native Meetup vom 28. August 2018 haben wir über das Thema Kubernetes Distribution berichtet, die Slides dazu sind auf Speaker Deck zu finden. Mehr zum Thema OpenShift, Docker und Kubernetes findest du ausserdem hier. Ein weiterer empfehlenswerter Blogpost zu diesem Thema von Tomasz Cholewa: 10 most important differences between OpenShift and Kubernetes (englisch, technisch).

Wie können wir helfen?

Durch unsere Erfahrung im Betrieb von OpenShift Clustern rund um die Welt bieten wir Managed OpenShift Cluster auf nahezu jeder Public, Private oder On-Premise Cloud an. Oder besteht Interesse an einer anderen Kubernetes Distribution als OpenShift? Wir helfen gerne bei der Evaluation, Integration und Betrieb und unterstützen mit unserer langjährigen Kubernetes Erfahrung.
Kontaktiere uns, folge uns auf Twitter oder wirf einen Blick auf unsere Services.
Wir freuen uns auf dein Feedback!