Broadcom’s Open Hardware Ecosystem Enables Flexibility and Choice for Modern Private Cloud Deployments

As the technology market continues to evolve, we see two new challenges in the near to medium term for customers: hardware costs and AI demand. To address these significant challenges, we’re expanding our open hardware certification program to increase OEM and ODM participation through new ODM self-service certification and VMware Cloud Foundation (VCF) AI ReadyNodes.

Our Vision for Extending the ReadyNode Program to ODM Hardware

Multi-vendor support for VCF clusters is Broadcom’s next step toward a more open and flexible hardware ecosystem and driving down costs. We’re also addressing the management challenges of multi-vendor support in VCF 9.0. Our most recent release enables vSphere Lifecycle Manager (vLCM) to update mixed-vendor clusters and supports multiple hardware support managers (HSMs) per cluster. Clusters can have up to four additional image definitions.

Looking forward, we have a broad vision and aggressive roadmap to solve the challenges associated with open ecosystems. Customers should not have to choose between cost savings and operational simplicity. Broadcom’s goal is to de-risk customers’ hardware of choice. As a first step, at Explore on Tour in Frankfurt, we announced the expansion of the VCF ReadyNode Program to enable ODM partners to self-certify ReadyNodes through the Broadcom Technology Alliance Program (TAP). All certified systems will be validated for full interoperability with VMware Cloud Foundation with consistent VCF lifecycle management.

This shift is impactful to many customers and providers, including OVHcloud, a global cloud player and the European Cloud leader, which delivers multiple VCF-based cloud solutions. According to an interview in February 2026 with OVHcloud, they believe this new program will accelerate their time-to-market by 30-50%, as they are able to self-certify more configurations, more frequently. OVHcloud operates its own data centers, designs and manufactures its own servers, and develops its own software platforms. To them, this vertical integration delivers competitive advantage, as they control every decision from infrastructure planning to customer delivery. Previously, any server design change would require external validation, which could take weeks. This new process provides OVHcloud faster access to certified infrastructure, more configuration options, reduced procurement friction and lower costs.

This certification process helps ensure that when customers choose a certified ODM solution, they are getting a platform that is pre-validated for seamless deployment, interoperability, and simplified lifecycle updates. They gain the ability to confidently deploy hardware from a diverse and competitive ecosystem, capturing CapEx savings—in the case of OVHcloud, potentially up to 50% on hardware.

The AI Solution: Addressing the Parallel AI Infrastructure Challenge with AI ReadyNodes

The urgency for a standardized, open infrastructure ecosystem is not limited to general compute and storage; it is mirroring the exact same trajectory in the AI market. As organizations race to adopt AI, they face a familiar set of hurdles: ballooning costs, vendor lock-in, and deployment complexity.

Broadcom is taking a step in the same direction as our VCF ReadyNodes with the introduction of AI ReadyNodes—an open, standards-based platform approach. Just as ODM ReadyNodes solve the hardware cost problem, the goal here is to eliminate the complexity of AI adoption while helping ensure interoperability.

AI ReadyNodes represent a unified solution stack that integrates VMware Cloud Foundation (VCF), NVIDIA AI Enterprise software, and certified hardware components. Customers get the assurance of certified hardware across the entire AI solution stack while benefitting from simplified lifecycle management and full VCF interoperability. Additionally, certified GPUs and integration with NVIDIA AI Enterprise software optimize performance for GPU-accelerated workloads, model fine-tuning, and inference workloads. This approach gives customers a clear migration path to fully optimized GPU solutions without the guesswork, allowing IT teams to accelerate their time to value and adopt new platforms faster.

As one of the first partners to participate, Supermicro is certifying its industry-leading GPU systems to ensure full interoperability with VMware Cloud Foundation. This collaboration offers customers a cost-effective, validated migration path for AI model training and inference that significantly reduces the complexity of deploying private AI infrastructure.

A Unified Future of Choice

Whether for private cloud infrastructure or the new wave of AI, the future is built on certified, standardized infrastructure that prioritizes choice and flexibility. By leveraging the open hardware ecosystem—through ODM partnerships and AI ReadyNodes—organizations can confidently deploy diverse hardware while maintaining the operational stability and financial control they demand.