SUSE Harvester
SUSE Harvester: open source HCI - traditional virtualization alternative. Kubernetes-native virtualization, Rancher integration. 100% open source, no vendor lock-in.

Key Features
- Open Source HCI - fully open source, no licenses
- Kubernetes-native - VMs as Kubernetes workloads
- Rancher Integration - unified management
- Longhorn Storage - software-defined storage
- Traditional virtualization alternative - VM migration
Table of Contents
What is SUSE Harvester?
Moving to Kubernetes-based virtualization moves the security boundary: alongside the hypervisor you now have to protect the cluster control plane, because that is what grants access to every machine at once. We cover the consequences further down.
SUSE Harvester is an open source hyper-converged infrastructure platform that runs virtual machines as Kubernetes workloads, on ordinary server hardware, with storage and management included.
Key Functions:
- Kubernetes-native - VMs as Kubernetes workloads
- Rancher Integration - management via Rancher
- Longhorn Storage - built-in software-defined storage
- Fully open source - zero licensing costs
What problem does it solve?
flowchart LR
subgraph Proprietary Virtualization Problem
B --> C[Vendor lock-in]
C --> D[Broadcom acquisition]
D --> E[Rising costs]
end
subgraph Harvester Solution
F[Harvester HCI] --> G[100% Open Source]
G --> H[Zero licenses]
H --> I[Kubernetes-native]
I --> J[Future-proof]
end
style A fill:#dc2626,stroke:#b91c1c,color:#fff
style B fill:#dc2626,stroke:#b91c1c,color:#fff
style C fill:#dc2626,stroke:#b91c1c,color:#fff
style D fill:#dc2626,stroke:#b91c1c,color:#fff
style E fill:#dc2626,stroke:#b91c1c,color:#fff
style F fill:#22c55e,stroke:#16a34a,color:#fff
style G fill:#22c55e,stroke:#16a34a,color:#fff
style H fill:#22c55e,stroke:#16a34a,color:#fff
style I fill:#22c55e,stroke:#16a34a,color:#fff
style J fill:#22c55e,stroke:#16a34a,color:#fff
Common Problems:
- Broadcom acquisition = rising virtualization licensing costs
- Vendor lock-in to proprietary virtualization
- Separate tools for VMs and containers
- Expensive HCI solutions (vSAN, Nutanix)
- No Kubernetes-native virtualization
Harvester Architecture
flowchart TD
A[Rancher Management] --> B[Harvester Cluster]
B --> C[Node 1]
B --> D[Node 2]
B --> E[Node 3]
C --> F[Kubernetes]
D --> F
E --> F
F --> G[KubeVirt - VM]
F --> H[Longhorn - Storage]
F --> I[Containers]
style A fill:#f59e0b,stroke:#d97706,color:#fff
style B fill:#8b5cf6,stroke:#7c3aed,color:#fff
style C fill:#6366f1,stroke:#4f46e5,color:#fff
style D fill:#6366f1,stroke:#4f46e5,color:#fff
style E fill:#6366f1,stroke:#4f46e5,color:#fff
style F fill:#22c55e,stroke:#16a34a,color:#fff
style G fill:#22c55e,stroke:#16a34a,color:#fff
style H fill:#22c55e,stroke:#16a34a,color:#fff
style I fill:#22c55e,stroke:#16a34a,color:#fff
Key Features
Kubernetes-native VMs
VMs as K8s workloads
- KubeVirt for VMs
- VMs as CRD in Kubernetes
- kubectl for VM management
- Live migration
Longhorn Storage
Built-in SDS
- Replicated block storage
- Snapshots and backup
- Disaster recovery
- Zero vSAN costs
Rancher Integration
Unified management
- Import to Rancher
- Multi-cluster view
- RBAC and SSO
- Monitoring
VM Templates
Fast deployment
- Cloud-init support
- Windows and Linux VMs
- GPU passthrough
- PCI passthrough
Networking
VM networking
- VLAN support
- Bridge networking
- SR-IOV (optional)
- Load balancer
Virtualization Migration
VM import
- Import VMDK
- OVA import
- Bulk migration
- VM conversion
Harvester vs vSphere vs Proxmox
| Aspect | Harvester | vSphere | Proxmox VE |
|---|---|---|---|
| License | Open Source (Free) | Enterprise (expensive) | Open Source + Support |
| Kubernetes | Native (is K8s) | vSphere + Tanzu | None |
| Storage | Longhorn (built-in) | vSAN (additional license) | Ceph/ZFS |
| Container management | Native | Tanzu (additional) | LXC |
| Rancher integration | Native | None | None |
| Enterprise support | SUSE subscription | Broadcom license | Proxmox subscription |
| GPU passthrough | Yes | Yes | Yes |
| Live migration | Yes | Yes | Yes |
| Use case | Modern workloads, K8s-first | Traditional enterprise | SMB, homelab |
Harvester Advantage:
- Fully open source - zero licensing costs
- Kubernetes-native - VMs and containers on one platform
- Longhorn built-in - no additional storage costs
- Rancher integration - one tool for VMs and K8s
- Edge-ready - lightweight deployment
Who is it for?
Harvester MAKES sense when:
- • Looking for traditional virtualization alternative (Broadcom)
- • Want VMs + Kubernetes on one platform
- • Edge computing - lightweight HCI
- • Have Rancher - native integration
- • Want to avoid licensing costs
Harvester DOESN'T make sense when:
- • Only legacy Windows VMs - vSphere better
- • "Big vendor" requirement from corporation
- • No Kubernetes experience
- • Very large environments (1000+ VMs) - young product
Specification
| Parameter | Value |
|---|---|
| Minimum nodes | 3 (HA), 1 (single node) |
| OS | SUSE Linux (immutable) |
| Hypervisor | KubeVirt |
| Storage | Longhorn |
| Networking | Canal, Multus |
| GPU | NVIDIA GPU passthrough |
| Management | Web UI, kubectl, Rancher |
| License | Apache 2.0 (Open Source) |
The attack surface of Kubernetes-based HCI
Harvester changes more than the licensing model — it changes what you have to protect. In classic virtualization the security boundary is the hypervisor. Here the Kubernetes control plane sits next to it, and it is often the shortest path to every virtual machine at once.
What to address during deployment:
- Access to the management layer. An account with rights in Rancher or directly in the cluster API can start, clone or delete any VM. RBAC and multi-factor authentication for the console carry more weight here than hardening any single machine.
- Network isolation between workloads. Default in-cluster communication tends to be too permissive. Network policies restrict east–west traffic — exactly the traffic an attacker uses after compromising one machine.
- Images and the supply chain. Kubernetes-native means some components come from external registries. Scanning images for vulnerabilities and controlling their sources is basic hygiene, not an option.
- Data protection in Longhorn. Replication protects against disk failure, not against a volume being encrypted. Backups must be separated from the cluster.
We verify management-layer configuration and segmentation as part of security audits, and environment resilience through penetration testing.
FAQ
Is Harvester a traditional virtualization replacement? Yes for modern workloads. For legacy Windows apps vSphere may be better, but Harvester supports Windows VMs.
What does traditional virtualization migration look like? Import VMDK/OVA, bulk migration tools. Harvester can import VMs from vSphere.
Do I need Rancher? Not required, but recommended. Rancher adds multi-cluster management, RBAC, monitoring.
What about performance vs traditional virtualization? Comparable for most workloads. KubeVirt adds minimal overhead.
How does support work? Open source is free. Enterprise support via SUSE subscription.
Does nFlo deploy Harvester? Yes. Greenfield deployments and traditional virtualization migrations. Rancher integration, storage and networking configuration.
Inquire about SUSE Harvester
Contact your product specialist and get a custom quote.

Related Services
Our services supporting the implementation and management of this solution
AIOps - AI for IT Operations
IT Infrastructure
Stop fighting fires. AIOps predicts problems before they impact business.
IT Infrastructure Management Automation
IT Infrastructure
Save 500+ hours per year by eliminating manual tasks. Ansible, Terraform, PowerShell.
Backup and Disaster Recovery
IT Infrastructure
Recover data in minutes, not days. Backup and DR with tested RTO/RPO.
Comprehensive Backup Services
IT Infrastructure
Never lose data again. Backup with tested recovery and ransomware protection.
From Our Knowledge Base
Articles related to this solution
How to choose an IBM Power server: a guide to Power11 and Power10 models
A guide to the current IBM Power family: how Power11 differs from Power10, when S1112 or S1012 is enough, how S1122, L1122 and S1124 really differ, when you need an E1150 or an E1180, and what to check in a quote.
On-premises infrastructure as a service: when the HPE GreenLake consumption model makes sense
Infrastructure as a service (IaaS) is a service model, not a place. We explain how the consumption model works in your own server room using HPE GreenLake as the example, when it is cheaper than buying, who is responsible for security and how to read the contract.
Immutable backup for businesses: why immutability is a property of the whole arrangement, not of the medium
A backup stops protecting you against ransomware the moment an attacker with administrative rights can delete it. Immutability is not a storage-array feature — it is a property of the whole arrangement: the medium, the authentication domain and the recovery procedure. This text shows where that arrangement usually breaks and how to check it.
Related Products
Other solutions you might be interested in
AdRem NetCrunch
AdRem Software
AdRem NetCrunch: agentless network and IT infrastructure monitoring. 680+ monitoring targets, automatic L2 topology maps.
baramundi Management Suite
baramundi
baramundi Management Suite: UEM platform with patch management for 200+ applications. IT automation, MDM, inventory. German manufacturer.
baramundi Proactive Hub
baramundi
baramundi Proactive Hub: a cloud platform for digital employee experience analysis, vulnerability risk prioritisation and endpoint remediation.
Citrix DaaS
Citrix
Citrix DaaS: Desktop as a Service with cloud management. HDX protocol, multi-cloud resources, auto-scaling, subscription model.
Want to Reduce IT Risk and Costs?
Book a free consultation - we respond within 24h
Or download free guide:
E-book: Digital Resilience Strategy 2026