Nutanix has announced that its Unified Storage (NUS) solution has received enterprise-level certification from NVIDIA, marking a significant development in hybrid multicloud computing.
The NVIDIA-Certified Storage designation assures enterprises and cloud providers that the solution meets the necessary performance, security, and scalability needed for extensive AI workloads in large-scale production environments. Nutanix also plans to enhance AI-native storage by incorporating support for NVIDIA Vera BlueField-4 STX, aiming to advance data access speeds, storage efficiency, and streamlined AI operations.
Enhancing GPU Utilization
As companies strive to construct robust AI infrastructures to manage production workloads, they require systems that can efficiently facilitate data movement, optimize GPU utilization, and mitigate deployment risks. This necessitates overcoming challenges like fragmented infrastructure and data silos, which hamper deployment efficiency and GPU performance, complicating AI scaling efforts.
Nutanix provides validated configurations to boost AI infrastructure deployment
With this certification, Nutanix provides validated configurations to boost AI infrastructure deployment within enterprises. By ensuring full-stack interoperability with NVIDIA technology, it significantly reduces I/O bottlenecks and integration risks, allowing for linear scalability to meet the data demands essential for optimal GPU and data use in production environments.
Addressing Modern AI Workload Requirements
Thomas Cornely, Executive Vice President of Product Management at Nutanix, commented, "To build and run AI factories successfully, enterprises must move past fragmented infrastructure and data silos that limit GPU infrastructure efficiency."
He emphasized the value of the NVIDIA certification in verifying that Nutanix Unified Storage offers full-stack interoperability, scalability, and reliable data flow required by contemporary AI tasks. Cornely highlighted the partnership with NVIDIA as a means to provide customers with a unified, high-performance platform to confidently scale AI operations.
Optimizing Storage for Agentic AI
Jason Hardy, Vice President of Storage Technology at NVIDIA, stated, "As enterprises scale their AI factory deployments to meet demanding agentic AI workloads, storage is foundational to unlocking full-stack performance, efficiency, and accuracy."
The newly certified NUS offers a reliable and interoperable base for eliminating data bottlenecks and ensures GPUs operate at peak efficiency, facilitating the confident scaling of AI workloads. Utilizing a 10-node, all NVMe cluster, NUS integrates advanced parallel NFS and GPUDirect Storage using NFS with RDMA to create an efficient, low-latency data path between GPUs and storage, thus maximizing uptime and resource utilization.
Achieving Large-Scale AI Performance
The NVIDIA-Certified Nutanix Unified Storage reference architecture is currently available
The architecture allows enterprises to transition from limited GPU deployments to expansive production settings, keeping storage performance consistent as AI workloads grow. To support large-scale AI demands, the NUS solution incorporates NVIDIA Spectrum X Ethernet, featuring NVIDIA Spectrum 4 switches and BlueField 3 DPUs, allowing data throughput scalability from 10 GB/s to 160 GB/s reads, and 5 GB/s to 80 GB/s writes across various GPU configurations.
This resilient architecture supports diverse AI workload requirements, from training to inference, across multiple compute platforms such as x86-based systems and NVIDIA HGX servers with specific GPU models.
The NVIDIA-Certified Nutanix Unified Storage reference architecture is currently available. Upcoming support for NVIDIA BlueField-4 STX is anticipated by mid-2026.
Nutanix, a pioneer in hybrid multicloud computing, announces the Nutanix Unified Storage (NUS) solution is NVIDIA-Certified at the enterprise level.
NVIDIA-Certified Storage is designed to enable enterprises and cloud providers to confidently deploy storage solutions that support the performance, security, and scale required for large-scale production AI workloads. Nutanix is also advancing AI-native storage with planned support for NVIDIA Vera BlueField-4 STX, reinforcing its focus on faster data access, greater storage efficiency, and simpler AI operations at scale.
Maximize GPU utilization
As enterprises and cloud providers race to build AI factories to support production AI workloads, they require infrastructure that can keep data moving, maximize GPU utilization, and reduce deployment risk. Success depends not only on access to powerful GPUs but on the ability to feed those systems with data efficiently and reliably. Fragmented infrastructure, siloed data, and inconsistent performance can slow deployments, limit GPU efficiency, and make AI harder to scale reliably.
With this certification, Nutanix is providing enterprises and cloud providers with a validated configuration to support enterprise deployment of AI infrastructure. The certification helps ensure NUS is validated for full-stack interoperability with NVIDIA AI infrastructure, helping to reduce I/O bottlenecks and integration risk. By enabling linear scalability for the data-hungry demands of AI workloads, it helps ensure an organization’s most valuable assets, its GPUs and data, are working at maximum efficiency in production environments.
Modern AI workloads
“To build and run AI factories successfully, enterprises must move past fragmented infrastructure and data silos that limit GPU infrastructure efficiency,” said Thomas Cornely, executive vice president, Product Management, Nutanix.
“This NVIDIA certification validates that Nutanix Unified Storage delivers the full-stack interoperability, linear scalability, and reliable data velocity that modern AI workloads demand. By collaborating closely with NVIDIA, we are giving customers a unified, high-performance foundation to scale their production AI operations with confidence.”
Agentic AI workloads
"As enterprises scale their AI factory deployments to meet demanding agentic AI workloads, storage is foundational to unlocking full-stack performance, efficiency, and accuracy,” said Jason Hardy, vice president, Storage Technology, NVIDIA. “Nutanix Unified Storage achieving NVIDIA certification gives customers a trusted, interoperable foundation to eliminate data bottlenecks, maximize GPU utilization, and scale production AI workloads with confidence."
Built on a 10-node, all NVMe cluster, NUS leverages enhanced parallel NFS (pNFS) and GPUDirect Storage over NFS with RDMA to establish a low-latency, high-throughput, and resilient data path directly between GPUs and storage—maximising utilization while minimizing downtime.
Large-scale AI performance
The result is a scalable foundation for enterprise AI that helps customers move from targeted GPU deployments to larger production environments while keeping storage performance predictable as AI workloads expand. To support large-scale AI performance, the solution uses NVIDIA Spectrum X Ethernet, including NVIDIA Spectrum 4 switches and BlueField 3 DPUs, and delivers linear scalability from 10 GB/s read and 5 GB/s write for 32 GPUs to 160 GB/s read and 80 GB/s write for 1,024 GPUs.
This resilient, zero-downtime architecture provides a flexible foundation for AI workloads, supporting training, fine-tuning, inference, and RAG pipelines across a wide range of compute platforms including x86-based systems (NVIDIA RTX 6000 PRO Blackwell, NVIDIA H200 NVL), NVIDIA HGX servers with B200, H200, or H100 GPUs, and NVIDIA GH200 Grace Hopper Superchip configurations.
The NVIDIA-Certified Nutanix Unified Storage reference architecture is available today. Planned support for NVIDIA BlueField-4 STX is expected to be available in the second half of 2026.
