Broadcom’s latest VCF 9.1 release delivers a unified private cloud platform that reduces AI infrastructure costs, strengthens security, and accelerates enterprise-scale deployment of inference and agentic workloads.
Broadcom Inc. has introduced VMware Cloud Foundation (VCF) 9.1, a major update designed to help enterprises deploy and manage production-grade artificial intelligence workloads with greater efficiency, stronger security, and significantly reduced costs. As organizations accelerate their adoption of AI—particularly inference and emerging agentic AI applications—VCF 9.1 positions itself as a unified private cloud platform purpose-built to address the operational, financial, and governance challenges associated with scaling AI in real-world environments.
At its core, VCF 9.1 delivers an AI- and Kubernetes-native private cloud architecture that integrates compute, storage, networking, and security into a single platform. It supports heterogeneous infrastructure across leading hardware ecosystems, including AMD, Intel, and NVIDIA, giving enterprises the flexibility to choose the most suitable GPUs and CPUs for their workloads. This open hardware approach is particularly critical at a time when supply constraints and cost pressures are forcing organizations to rethink rigid infrastructure strategies.
A preview of Broadcom’s Private Cloud Outlook 2026 report underscores a broader industry shift: private cloud environments are increasingly becoming the preferred foundation for production AI. According to the findings, 56% of organizations are already running—or planning to run—inference workloads in private cloud settings, compared to 41% relying on public cloud, a figure that has declined notably year over year. Cost concerns are a dominant factor, with 62% of IT leaders expressing significant anxiety over generative AI infrastructure expenses. Additionally, 36% highlight new requirements around data protection, privacy, and regulatory compliance as AI adoption expands.
VCF 9.1 addresses these concerns by enabling enterprises to maximize the value of their existing infrastructure rather than relying on costly expansions or unpredictable public cloud consumption models. Through intelligent resource optimization techniques, the platform can reduce server costs by up to 40%, largely driven by advanced memory tiering that allows mixed AI and non-AI workloads to coexist efficiently. Storage costs are also significantly lowered—by as much as 39%—through enhanced compression and deduplication technologies tailored for AI data pipelines. On the operational side, Kubernetes management costs can be reduced by up to 46%, while infrastructure scalability improves with faster upgrades and increased fleet capacity.
One of the defining advantages of VCF 9.1 is its ability to consolidate diverse workloads onto a single platform. Enterprises can run traditional virtual machines, containerized applications, AI inference services, and agentic workflows without maintaining separate infrastructure stacks. This consolidation eliminates operational silos and reduces the complexity typically associated with hybrid environments. It also accelerates application delivery, allowing development teams to move from experimentation to production more quickly.
The platform’s Kubernetes capabilities have been significantly enhanced, offering greater scalability, faster deployment times, and reduced upgrade windows. These improvements are critical for AI-driven applications that demand high availability and rapid iteration cycles. Additionally, VCF 9.1 introduces mixed compute management, enabling organizations to efficiently allocate CPU and GPU resources based on workload requirements. This is particularly important for agentic AI, which often relies heavily on CPU resources for orchestration and decision-making, alongside GPU acceleration for inference.
Beyond performance and cost optimization, VCF 9.1 places a strong emphasis on observability and governance. Enterprises gain access to detailed metrics such as time-to-first-token, token throughput, and GPU utilization across multiple accelerator types. These insights enable precise resource allocation and help maximize return on investment. At the same time, centralized policy controls ensure compliance with data sovereignty and regulatory requirements, which are increasingly critical in AI deployments involving sensitive data.
Security is another foundational pillar of VCF 9.1. The platform adopts a zero-trust architecture that integrates protection mechanisms directly into the infrastructure layer, rather than relying on external tools. This approach ensures that AI models, training data, and applications are secured from the hypervisor level upward. Features such as zero-trust segmentation and distributed threat detection extend protection across both virtual machines and Kubernetes workloads, providing comprehensive coverage for modern AI environments.
VCF 9.1 also introduces advanced ransomware recovery capabilities, including isolated recovery environments and integrated validation tools. These features are designed to safeguard high-value AI assets—such as proprietary models and datasets—while minimizing downtime and avoiding costly data transfers during recovery scenarios. Continuous compliance enforcement further strengthens the platform’s security posture by automating regulatory checks and maintaining audit readiness without manual intervention.
Another notable innovation is zero-downtime live patching, which allows infrastructure updates to be applied without disrupting running workloads. This capability is particularly valuable for AI applications that require continuous availability, such as real-time inference services or mission-critical automation systems. By eliminating maintenance windows, organizations can maintain service-level agreements while keeping their environments secure and up to date.
Networking performance has also been enhanced to meet the demands of large-scale AI workloads. Support for high-speed network interfaces and advanced data transfer technologies enables efficient multi-host training and rapid data movement, both of which are essential for generative AI and large language models. Combined with virtualized load balancing and security services, VCF 9.1 reduces the need for dedicated hardware appliances, further lowering capital expenditures.
The platform’s open ecosystem is reinforced through partnerships with leading technology providers. Integration with advanced GPU architectures and high-performance networking solutions ensures that enterprises can achieve the same level of performance as public cloud environments—while maintaining full control over their data and infrastructure. This balance between performance and sovereignty is becoming increasingly important as organizations seek to protect intellectual property and comply with regional regulations.
Customer and partner feedback highlights the practical benefits of this approach. Enterprises deploying AI workloads on private cloud infrastructure report improved cost predictability, enhanced data security, and greater operational efficiency. By leveraging existing infrastructure investments, organizations can avoid the financial and logistical challenges associated with large-scale hardware procurement or public cloud dependency.
From an operational standpoint, VCF 9.1 introduces automation capabilities that significantly reduce administrative overhead. Fleet management can scale to thousands of hosts, while automated upgrades and patching eliminate the need for manual intervention. This allows IT teams to focus on strategic initiatives rather than routine maintenance, improving overall productivity and agility.
The introduction of reusable application blueprints further accelerates deployment workflows. By capturing complex multi-VM configurations as templates, organizations can quickly replicate environments across development, testing, and production stages. This not only reduces deployment time but also ensures consistency and minimizes the risk of configuration errors.
In summary, VMware Cloud Foundation 9.1 represents a comprehensive solution for enterprises looking to operationalize AI at scale. By combining cost optimization, performance enhancements, integrated security, and hardware flexibility, the platform addresses the key challenges that have historically limited AI adoption in production environments. Its emphasis on private cloud infrastructure aligns with broader industry trends, as organizations seek greater control, predictability, and compliance in their AI strategies.
As AI continues to evolve—from basic inference to complex agentic systems—the need for robust, scalable, and secure infrastructure will only grow. VCF 9.1 positions itself as a critical enabler of this transformation, providing enterprises with the tools and capabilities required to turn AI from an experimental technology into a core business driver.
Source link: https://www.broadcom.com
