The most significant change introduced by Broadcom is not licensing, pricing, or even portfolio reduction. The real shift is that VMware Cloud Foundation is no longer optional in spirit, even when it is optional on paper. VCF has become the reference operating model for how VMware expects its platform to be deployed, upgraded, and maintained.
Before this shift, VMware environments were often built incrementally. vSphere might have been installed years earlier, NSX added later for a specific use case, and vSAN introduced only where it made sense. Each component lived on its own lifecycle, owned by different teams, upgraded at different times, and occasionally tuned outside vendor recommendations to meet local requirements.
VCF deliberately removes this flexibility.
At the center of this change is SDDC Manager. In a VCF-based environment, SDDC Manager is not a convenience tool; it is the authority. It defines which versions of ESXi, vCenter, NSX, and vSAN are allowed to exist together through the VCF Bill of Materials (BOM). Once an environment is under SDDC Manager control, individual components stop being independently managed products and become subsystems of a single platform lifecycle.
This has a direct technical impact on how infrastructure teams operate. An ESXi upgrade is no longer just an ESXi upgrade. It is a platform upgrade that must respect dependency order, firmware compatibility, driver versions, and NSX kernel module alignment. Attempting to bypass this process – by manually upgrading a host, patching NSX outside the VCF workflow, or introducing unsupported drivers – does not simply create risk; it actively blocks future lifecycle operations.
Workload domains reinforce this operating model. In VCF, the workload domain replaces the cluster as the primary unit of architecture. Each domain has its own vCenter, its own lifecycle cadence, and a clearly defined purpose. This allows strong isolation between platform services, production workloads, and security zones, but it also introduces additional operational overhead. Architects must plan domain boundaries carefully because moving workloads between domains later is not trivial.
NSX’s role inside VCF is also subtly but importantly different. In pre-VCF environments, NSX was often treated as an advanced feature, deployed only where micro segmentation or overlay networking was explicitly required. In VCF, NSX is foundational. Overlay networking, distributed firewalling, and centralized routing are assumed parts of the design, not optional enhancements. This shifts the responsibility of network and security design much earlier in the project lifecycle and forces closer collaboration between traditionally separate teams.
From an operational perspective, VCF exposes weak processes very quickly. Organizations accustomed to making undocumented changes, performing emergency fixes directly on hosts, or running production environments without a staging area will struggle. VCF is intolerant of configuration drift. Deviations from the validated state are not just unsupported; they are actively detected and penalized during upgrades.
This rigidity is not accidental. Broadcom’s VMware strategy clearly prioritizes environments where infrastructure is treated as a controlled platform rather than a collection of loosely governed components. The trade-off is clear: flexibility is reduced in exchange for predictability, repeatability, and supportability at scale.
For organizations prepared to adopt this mindset, VCF provides a strong and stable foundation. For those expecting VMware to remain a toolbox of independently tunable products, the experience will feel restrictive. The technology itself has not become worse – but the rules of engagement have changed.
