Skip to main content

Overview

AWS EKS Anywhere allows you to run Amazon EKS on your own infrastructure (on-premise, edge locations). Connect your EKS Anywhere cluster to Qovery using BYOK (Bring Your Own Kubernetes).

What is EKS Anywhere?

EKS Anywhere is a deployment option for Amazon EKS that lets you create and operate Kubernetes clusters on your own infrastructure:
  • On-premise data centers
  • Edge locations
  • Private clouds
  • Air-gapped environments

Features

https://mintcdn.com/qovery/bC94RbU5OE288_w9/images/logos/cloud-providers/aws-icon.svg?fit=max&auto=format&n=bC94RbU5OE288_w9&q=85&s=39fdefda6b2a184faa10bc1d85b46f7d

AWS Consistency

Same EKS experience as cloud-based EKS

Full Control

Deploy on your own hardware and network

AWS Support

AWS support and tooling available

Hybrid Cloud

Bridge on-premise and cloud workloads

Prerequisites

EKS Anywhere cluster (v0.15.0+)
Kubernetes 1.24+
kubectl admin access
Network connectivity to Qovery Control Plane
Load balancer solution (MetalLB, HAProxy, etc.)

Supported Infrastructure

  • VMware vSphere (most common)
  • Bare metal servers
  • Nutanix (coming soon)
  • Apache CloudStack

Setup

1

Deploy EKS Anywhere

Follow AWS documentation to create EKS Anywhere cluster:
eksctl anywhere create cluster \
  -f cluster-config.yaml
2

Install Prerequisites

  • Load balancer (MetalLB for bare metal)
  • Storage class (local-path, NFS, or enterprise solution)
  • Metrics server
3

Connect to Qovery

Follow BYOK setup guide to connect cluster to Qovery:BYOK Setup Guide
4

Deploy Applications

Start deploying applications on your on-premise cluster

Use Cases

Keep data on-premise for compliance or regulatory requirements while using Kubernetes
Run workloads closer to data sources or users for lower latency
Deploy in environments without internet access for security requirements
Bridge on-premise infrastructure with cloud deployments seamlessly

Comparison

FeatureEKS AnywhereCloud EKS
InfrastructureYour hardwareAWS-managed
Control PlaneYour controlAWS-managed
CostYour hardware costsPay-per-use
NetworkYour networkAWS VPC
UpdatesManualAutomated
SupportAWS Enterprise SupportIncluded

Best Practices

High Availability

  • 3+ control plane nodes
  • 3+ worker nodes
  • Spread across hardware/racks
  • External etcd backups

Networking

  • Dedicated VLAN for Kubernetes
  • Load balancer with HA
  • Firewall rules for Qovery
  • DNS resolution

Storage

  • Enterprise storage solution
  • Regular backups
  • Persistent volume provisioning
  • Snapshot capabilities

Monitoring

  • Deploy Prometheus/Grafana
  • Monitor hardware health
  • Track resource usage
  • Set up alerts

Next Steps