Hyper-V
High End
Consulting
Storage Spaces Direct ·
VMware replacement
Why the hypervisor is not a commodity
Virtualisierung galt lange als gelöst: ein Cluster, ein SAN, fertig. Dann kamen die Lizenzänderungen bei VMware, die Erkenntnis, dass die Public Cloud nicht für jede Last die richtige Antwort ist, und Ransomware-Gruppen, die gezielt den Hypervisor angreifen, weil dort alle Systeme auf einmal zu holen sind. Plötzlich ist die Plattformfrage wieder eine strategische Frage – und Hyper-V mit Storage Spaces Direct eine der wenigen Antworten, die Souveränität, Wirtschaftlichkeit und Enterprise-Reife verbinden.
Andy Wendel has worked with Hyper-V since its first release. In 2008 he was introduced to the platform by a Microsoft Premier Field Engineer, was trained over many years by Carsten Rachfahl – Microsoft MVP and Germany’s leading authority on Hyper-V and Storage Spaces Direct – and spent two weeks at the Enterprise Engineering Center in Redmond, the lab where Microsoft tests datacenter architectures with selected customers under real-world conditions. From this experience came the Masterclass Hyper-V and projects in which S2D clusters with RDMA, Switch Embedded Teaming and Network ATC were built from practice, not from the whitepaper.
The right choice: Andy Wendel
Hyper-V from the very beginning. On the platform since 2008, introduced by a Microsoft Premier Field Engineer.
Trained by Germany’s Hyper-V authority. Trained over many years by Carsten Rachfahl, Microsoft MVP for Hyper-V and Storage Spaces Direct.
Two weeks at the Enterprise Engineering Center in Redmond. Datacenter architectures tested under real-world conditions – where Microsoft validates its own reference designs.
Practice, not whitepapers. S2D clusters with RDMA, SET and Network ATC built and documented in [industry, insurance, public sector] – references on request.
Trainer of trainers. Author of the Masterclass Hyper-V; Microsoft Certified Trainer and one of 56 Microsoft Certified Learning Consultants worldwide.
Hypervisor, AD and PKI from a single source. The hypervisor is Tier 0. Someone who masters all three layers does not build a platform that endangers Active Directory.
Cleared for confidential work. Specialised in VS-NfD environments; secure collaboration via DRACOON.
Vendor-independent. No commission on servers, switches or storage. Recommendations are based solely on your environment.
Personal. You talk to the person who plans and delivers – no sales team, no handover to a junior crew.
„A cluster is only highly available once you have switched it off and brought it back up – before the customer does.“
Design and build of Hyper-V and S2D clusters
A hyper-converged cluster is not a hardware purchase followed by an installation, but an interplay of hardware selection, network design, storage layout and operating concept, in which every mistake endangers the availability of all workloads. We plan and build platforms that run from day one the way the datasheet promises.
Design principles
Validated hardware, configured correctly: Selection of servers, NVMe/SSD layout, RDMA-capable network adapters (RoCE or iWARP) and switches from the Windows Server and Azure Local catalogue – plus the firmware, driver and switch configuration that no datasheet mentions.
Networking to Microsoft reference: Switch Embedded Teaming (SET), Network ATC, Data Center Bridging, Priority Flow Control, separate intents for management, compute and storage – without packet loss under load.
Storage Spaces Direct: Resiliency choice (mirror, nested mirror, mirror-accelerated parity), cache layout, CSV design, ReFS, deduplication – tuned to your workloads instead of defaults.
Security by design: Shielded VMs and Host Guardian Service where needed, Secure Boot, TPM and Credential Guard on the hosts, Tier 0 isolation for domain controllers, hardened management paths – the hypervisor treated as a Tier 0 system.
Operability: Cluster-Aware Updating, monitoring (Windows Admin Center, SCVMM or third-party), backup integration, capacity planning and an operations manual that lets your team fix incidents themselves.
Scope of services
- Requirements and sizing workshop: Which workloads, which availability targets, what growth, which compliance requirements?
- Architecture document (hardware, network, storage, cluster topology, stretched cluster or replica, backup, security).
- Build of the platform including network and switch configuration, S2D enablement, cluster validation and load testing.
- Failover, outage and restart tests: pull network cables, power off a node, simulate storage failure – before the first production workload runs.
- Handover with operations manual and enablement – as a Masterclass Hyper-V for your team on request.
VMware replacement and migration to Hyper-V
For many organisations VMware has ceased to be an economical option since the licensing changes of recent years. Hyper-V is included in Windows Server Datacenter, Storage Spaces Direct replaces the SAN, and the features that matter in enterprise operations – live migration, replica, clustering, shielded VMs – have been mature for years. The migration itself is a project that needs planning: network and storage concepts differ, guest tools have to be swapped, backup and monitoring reconnected.
Typical scenarios
VMware vSphere to Hyper-V migration: Workload assessment, build of the target platform, conversion and move of VMs in waves, replacement of vCenter, NSX or vSAN with Windows Server built-in capabilities.
Cluster migration across domains and forests: Moving failover clusters and their workloads across domain boundaries – after mergers or an AD rebuild, for instance. A topic with almost no public documentation, for which we have developed our own runbooks.
Generation change: Replacing Windows Server 2012/2016/2019 clusters with Windows Server 2025 – via rolling cluster upgrade or rebuild, depending on what your environment allows.
Consolidation: Merging multiple clusters or sites, stretched clusters for datacenter redundancy, Hyper-V Replica for disaster recovery.
Cloud exit: Repatriating workloads from Azure, AWS or other public clouds onto your own platform – with a cost model, capacity planning and an operating concept that does not simply give up the advantages of the cloud.
Review, hardening and performance analysis
Many Hyper-V environments run – but not the way they could: storage latencies in the millisecond range because RDMA was never configured properly; live migrations that take minutes; clusters whose quorum configuration shuts everything down in an emergency; hosts on which domain admins log on interactively. A review uncovers what stays invisible day to day and decides everything during an outage.
Areas of review
Availability: Quorum, witness, cluster networks, storage redundancy, backup consistency, documented and tested restart procedures.
Performance: Storage path (RDMA, DCB, PFC, cache), CPU and memory overcommitment, NUMA, VM configuration, network throughput – measured, not estimated.
Security: Host hardening, administration paths, Tier 0 separation of domain controllers, backup access, ransomware resilience of the hypervisor, need for shielded VMs.
Operations: Patch process, monitoring, capacity reserves, documentation – and whether your team can cope alone at three in the morning.
Result: a prioritised list of measures, ordered by impact in your environment. Implementation on request.
Domain controllers on Hyper-V – done right
Virtualised domain controllers are standard, and so are the mistakes: snapshots that lead to USN rollbacks, backups that do not capture system state, hosts on which the domain admin is logged on and can therefore take over every VM. We design the virtualisation of Tier 0 systems so that hypervisor and Active Directory do not endanger each other – a topic that requires AD and Hyper-V expertise at the same time, and for which we have developed our own procedures for backup, snapshot and recovery.
Classified environments, compliance and sovereignty
For public authorities, the defence and security industry and operators of critical infrastructure, the hypervisor is the foundation of digital sovereignty: what runs on it stays in your own datacenter. Andy Wendel specialises in projects under the German classification level VS-NfD (“restricted”) and builds platforms that work without any cloud connection – isolated management networks, shielded VMs with Host Guardian Service, documented administration paths, project communication via German-operated platforms such as DRACOON. The designs meet the evidence requirements of ISO 27001, BSI IT-Grundschutz, NIS 2 and DORA.
How we work
1 · Assessment. Inventory of the existing virtualisation, workloads, networks, storage and dependencies – with measurements instead of assumptions.
2 · Design. Architecture document with sizing, network and storage concept that justifies decisions. Understandable for decision-makers, actionable for administrators.
3 · Proof. Build, validation and failure testing of the platform before the first production workload moves.
4 · Execution. Migration in waves, with a rollback path per workload and defined acceptance criteria – no surprises for the business units.
5 · Handover and enablement. Operations manual, emergency procedures and knowledge transfer – as a Masterclass Hyper-V on request. The goal is independence.
At eye level, with respect for what your team has already built – and with the clarity you are entitled to expect from an external expert.
Let’s talk about your Hyper-V needs
In a no-obligation 30-minute initial consultation, we’ll assess the current state of your platform, explore the available options – VMware replacement, a complete rebuild, moving away from the cloud or hardening – and determine whether we’re a good fit. No sales pitch – just a technical discussion.
This post is also available in:
