SUSE Rancher Prime: On-Prem Kubernetes Ledger

Part of the On-Prem Kubernetes Ledger (edition 2026-10-13). How the scores work: methodology.

1. Snapshot

Tier: Leader incumbent
Owner: SUSE S.A., owned by EQT private equity since the November 2023 delisting; Rancher Labs acquired December 2020
Licence model: Open source (Rancher repository Apache 2.0) with paid SUSE Rancher Prime subscriptions; prices not published
Products scored: Rancher Manager / Rancher Prime 2.x, RKE2, K3s, Fleet, Longhorn, NeuVector, Kubewarden, SUSE Virtualization (Harvester)
Latest release: Rancher v2.15.0 (2026-07-30; Prime 2.15.x GA 2026-08-27), RKE2 and K3s v1.37.0 (2026-09-14), Harvester 1.9 (2026-09-16)
Next signal: No dated public roadmap; Fleet deprecations announced in 2.15 notes, VM Auto Balance in early access
Archetype: Open multi-cluster manager
Evidence grade: A. Dates, releases, lifecycle and advisories come from the GitHub API and SUSE lifecycle page; scale claims, prices and some 2016 cells are unconfirmed.

2. Scorecard

Dimension	Score	Direction
Capability today	83.4 / 100	higher is better
Momentum since end of 2022	+8.8 points	higher is better
Credibility	20.0 / 25	higher is better
Commercial risk	8.5 / 25	lower is better
Technical lock-in	3.0 / 25	lower is better
Integrator fit	16.0 / 20	higher is better
AI leverage	13.5 / 20	higher is better
Ledger Index	77.5 / 100	balanced weights

3. Capability by domain

Scores 0 to 5 against fixed anchors; total is weighted to 100.

Domain (weight)	2016	2019	2022	Today
Install, upgrade and lifecycle (12)	2	3	3.5	4
Bare metal, edge and on-prem infrastructure (10)	2	3.5	4	4.5
Multi-cluster and fleet management (10)	1	3.5	4.5	4.5
Networking (10)	2	3	4	4.5
Storage and data protection (10)	1.5	2	4	4
Security and compliance (12)	1.5	3	4.5	4.5
Developer platform (8)	2	3	3	3.5
Observability and day-2 operations (8)	1.5	3	3.5	4
VMs on Kubernetes (VMware exit) (10)	0	0	3.5	4.5
GPU and accelerator workloads (10)	0	1	2.5	3.5
Total / 100	27.0	50.0	74.6	83.4

Year by year: 2016: 27.0 · 2017: 27.0 · 2018: 27.0 · 2019: 50.0 · 2020: 50.0 · 2021: 50.0 · 2022: 74.6 · 2023: 74.6 · 2024: 74.6 · 2025: 74.6 · 2026: 83.4

What moved the score

Year	Domain	Change	Points	Trigger
2019	Bare metal, edge and on-prem infrastructure	2 → 3.5	+3.0	RKE on any Linux host and node drivers, K3s 1.0 (Nov 2019) for edge and ARM, air-gap installs
2019	Developer platform	2 → 3	+1.6	Helm catalogs, multi-cluster apps and Rancher pipelines
2019	GPU and accelerator workloads	0 → 1	+2.0	Upstream device plugins only, no Rancher GPU tooling
2019	Install, upgrade and lifecycle	2 → 3	+2.4	Rancher 2.3 (Oct 2019): RKE in-place upgrades, new Kubernetes versions without Rancher upgrade, published windows
2019	Multi-cluster and fleet management	1 → 3.5	+5.0	Rancher 2.2/2.3: central RBAC over RKE, imported and hosted clusters, multi-cluster apps
2019	Networking	2 → 3	+2.0	RKE CNI choice (Canal, Calico, Flannel, Weave), network policy, ingress, Istio integration
2019	Observability and day-2 operations	1.5 → 3	+2.4	Prometheus cluster and project monitoring (2.2) with alerting and cluster logging
2019	Security and compliance	1.5 → 3	+3.6	Rancher 2.3 global, cluster and project RBAC with SSO providers, pod security policies
2019	Storage and data protection	1.5 → 2	+1.0	CSI and in-tree storage; Longhorn not yet GA (GA May 2020)
2022	Bare metal, edge and on-prem infrastructure	3.5 → 4	+1.0	Harvester 1.0/1.1 bare-metal HCI with ISO and PXE install, Elemental OS management 1.0 (Nov 2022)
2022	GPU and accelerator workloads	1 → 2.5	+3.0	NVIDIA GPU operator usable on RKE2, Harvester 1.1 GPU passthrough
2022	Install, upgrade and lifecycle	3 → 3.5	+1.2	RKE2 and K3s via Cluster API based provisioning (2.6), K3s provisioning GA in 2.7, upstream lag under a month
2022	Multi-cluster and fleet management	3.5 → 4.5	+2.0	Fleet GitOps bundled since 2.5 (Oct 2020), full AKS, GKE, EKS lifecycle, cluster templates
2022	Networking	3 → 4	+2.0	RKE2 bundles Canal, Cilium, Calico, Flannel and Multus; dual stack at cluster creation
2022	Observability and day-2 operations	3 → 3.5	+0.8	Monitoring v2 and Logging v2 since 2.5, Rancher backups
2022	Security and compliance	3 → 4.5	+3.6	RKE2 FIPS 140-2 and CIS defaults, DISA STIG for RKE2 (Nov 2022), NeuVector runtime security
2022	Storage and data protection	2 → 4	+4.0	Longhorn replicated block storage with snapshots and S3/NFS backup, Rancher backup operator
2022	VMs on Kubernetes (VMware exit)	0 → 3.5	+7.0	Harvester 1.1 (Oct 2022): KubeVirt VMs, live migration, snapshots, backup, VM import, managed in Rancher
2026	Bare metal, edge and on-prem infrastructure	4 → 4.5	+1.0	Rancher 2.12 (Jul 2025) official ARM64; vSphere, Nutanix, Harvester drivers; native CAPI providers GA 2.15
2026	Developer platform	3 → 3.5	+0.8	Prime Application Collection GA (Mar 2024), virtual clusters GA (Nov 2025); no portal or CI
2026	GPU and accelerator workloads	2.5 → 3.5	+2.0	RKE2 documents NVIDIA operator, runtime auto-detect (2024), Harvester MIG vGPU (1.7); NVIDIA only
2026	Install, upgrade and lifecycle	3.5 → 4	+1.2	Median 19-day lag 2023-2026, CAPI providers GA in 2.15 (Jul 2026); about 18 months per minor, no LTS
2026	Networking	4 → 4.5	+1.0	SR-IOV via Multus documented, Gateway API option (2.14, Mar 2026); Istio moved to Prime only
2026	Observability and day-2 operations	3.5 → 4	+0.8	SUSE Observability (StackState, acquired Jun 2024) for Prime customers, backup metrics (2.11)
2026	VMs on Kubernetes (VMware exit)	3.5 → 4.5	+2.0	SUSE Virtualization 1.8/1.9: vGPU and MIG vGPU, TPM, OVF import, storage live migration

4. Direction, 2023 to 2026

Each item carries one theme and one driver (V vision, C customer need, M market window, U upstream, P portfolio).

Date	Item	Theme	Driver
2023-12	Rancher 2.8: IPv6 RKE2 provisioning, Kubewarden replaces OPA Gatekeeper	D9 Security	U
2024-03	Rancher Prime 3.0 and Rancher Enterprise single package and price	D12 Packaging and licensing	P
2024-04	Rancher Turtles brings Cluster API GA to Rancher	D1 Install and lifecycle	U
2024-06	StackState acquisition for SUSE Observability in Rancher Prime	D8 Observability and operations	P
2024-11	SUSE AI made available on Rancher Prime	D13 Accelerators	M
2025-07	RKE1 end of life; ARM64 support and HelmOps GA in Rancher 2.12	D1 Install and lifecycle	C
2025-11	Liz AI assistant preview and virtual clusters GA	D7 AI-assisted operations	M
2025-12	Harvester 1.7: MIG-backed vGPU and OVF import	D10 VMs on Kubernetes	C
2026-03	Liz GA as multi-agent assistant with Rancher MCP server	D7 AI-assisted operations	V
2026-04	Harvester 1.8 storage live migration	D10 VMs on Kubernetes	C
2026-07	Rancher 2.15: native Cluster API providers GA	D2 Bare metal, edge and infrastructure	U

Centre of gravity: Install and lifecycle (2), AI-assisted operations (2), VMs on Kubernetes (2)
Driver mix: Vision 9%, Customer need 27%, Market window 18%, Upstream 27%, Portfolio 18%

5. Credibility: 20.0 / 25

Criterion	Score	Basis
Cadence	4.5	Rancher minors about every four months since 2.9 and monthly RKE2/K3s patches, dates published on SUSE lifecycle page
Roadmap transparency	4	Public GitHub issues, milestones and detailed release notes; no written multi-year roadmap found
Say-do	4.5	7 of 7 verifiable promises 2023-2026 delivered on time or within five months; 2 unconfirmed, 2 pending
Velocity	4	Several substantive features per minor, for example Turtles, IPv6, ARM64, CAPI providers, Harvester vGPU
Lifecycle stability	3	About 18 months per Rancher minor, no LTS; RKE1 end of life July 2025 forced replatforming to RKE2

Say-do record, 2023 to 2026: 11 items tracked; 7 delivered within 12 months with both ends confirmed (CAPI GA April 2024, SUSE AI, RKE1 end of life, Istio deprecation, Liz GA, CAPI providers GA, storage live migration); StackState GA date and Rancher Prime SaaS launch unconfirmed; MCP proxy and VM Auto Balance pending. 12 of 16 pass-1 items removed as unsupported.

6. Commercial risk: 8.5 / 25 (lower is better)

Criterion	Score	Basis
Price and licence volatility	1.5	Rancher Enterprise single package and price (Mar 2024); Istio moved to Prime-only Application Collection (2.12)
Purchase constraints	1	Free community edition of every component; Prime sold to any size, prices not published
Channel and access	1	Open downloads, public docs and GitHub, open SUSE One partner programme
Owner stability	3	EQT private equity owner since the November 2023 take-private
Cost of staying supported	2	Free community updates, but Prime minors supported about 18 months and some builds Prime-only

Security record of the product itself: No CISA KEV entry for Rancher, RKE2, K3s, Fleet or Harvester. 14 critical advisories 2023-2026, e.g. CVE-2023-22651, CVE-2024-22036, CVE-2025-62877, CVE-2026-44939; fixes shipped with disclosure (the security record).

7. Technical lock-in: 3.0 / 25 (lower is better)

Criterion	Score	Basis
Formats	0.5	CNCF-certified RKE2 and K3s every minor; Rancher CRDs live only in the management cluster
Export path	1	Downstream clusters keep running when detached; standard Velero-style backups apply
Stack coupling	0.5	Choice of Linux OS, five bundled upstream CNIs, any CSI driver
Hardware / cloud coupling	0	Any hardware, fully air-gapped installs, no mandatory SaaS
Skills and tooling coupling	1	Standard kubectl, Helm and GitOps skills; Rancher UI and Fleet add some own concepts

8. Integrator fit: 16.0 / 20

Criterion	Score	Basis
Partner programme openness	4	SUSE One Partner Program with open application and six specialisations
Multi-tenancy and self-service	4	Projects with RBAC and quotas, virtual clusters GA (Nov 2025), Harvester multi-tenancy via Rancher
API and infrastructure-as-code	4.5	Official rancher2 Terraform provider (54.8 million downloads), Rancher Turtles and CAPI RKE2 provider, Helm
Skills and certification	3.5	SUSE exams for Rancher Prime and SUSE Virtualization; very large K3s and Rancher community

9. AI leverage: 13.5 / 20

Strand	Today	Latest step
AI workload support	1.5	2025: SUSE AI adds an inference engine on Rancher Prime (Nov 2025)
AI platform stack	3.5	2025: SUSE AI with inference engine, observability and MCP proxy preview (Nov 2025)
AI-assisted operations	4.5	2026: Liz GA (Mar 2026): multi-agent assistant that can update resources, human approval for actions
Agent openness	4	2026: Rancher MCP server 1.0 (Mar 2026) acting with the user's Rancher token and RBAC; external MCP servers

AI say-do: 4 announced, 2 GA (SUSE AI, Liz), 1 preview pending (MCP proxy), 1 not confirmed (MLflow agents). Liz reads cluster state through the Rancher MCP server under the user's permissions and can propose and, with approval, apply changes.

10. Reading

SUSE is turning Rancher into a Cluster API based manager for any Kubernetes plus KubeVirt virtualization and an AI assistant for operators. The record is strong on shipping: upstream minors land in RKE2 or K3s within about three weeks, and the few dated promises found were kept. The trade-off is short 18-month support per Rancher minor, a private-equity owner, and a long list of critical management-plane advisories that need prompt patching.

11. Open questions and evidence caveats

Rancher Prime prices are not public, so commercial risk rests on packaging news, not price data.
Fleet scale claims and SUSE Observability GA date were not confirmed from opened sources.
Liz GA status rests on a trade report and the 1.0 tag; SUSE's own GA statement was not opened.
14 critical advisories in Rancher, Fleet and Harvester 2023-2026; none exploited per KEV, but patch cadence matters.