At COMPUTEX in Taipei, Taiwan, NVIDIA has launched a complete initiative to drive the worldwide transformation of enterprise IT infrastructure by means of its RTX PRO Servers and Enterprise AI Manufacturing facility validated design. This strategic transfer goals to switch conventional CPU-based programs with GPU-accelerated knowledge facilities that help a variety of demanding workloads, together with synthetic intelligence, engineering simulation, and real-time digital twin purposes.
On the heart of this initiative is the NVIDIA RTX PRO 6000 Blackwell Server Version GPU. Based mostly on the brand new Blackwell structure, the GPU delivers enhanced efficiency and power effectivity for enterprise environments. It’s designed to deal with a broad combine of information heart workloads, serving to corporations transition towards infrastructure optimized for AI and superior computing duties.
These new servers are a core element of NVIDIA’s Enterprise AI Manufacturing facility blueprint, which is a full-stack resolution that features {hardware}, networking, storage, and software program integration. The platform combines RTX PRO Servers with NVIDIA BlueField knowledge processing models (DPUs), Spectrum-X Ethernet networking, NVIDIA-Licensed Storage programs, and the NVIDIA AI Enterprise software program suite. This configuration helps the deployment of on-premises AI factories – knowledge facilities that allow companies to automate operations, construct AI-powered merchandise, and improve workforce productiveness with clever programs.
Jensen Huang, founder and CEO of NVIDIA, said that AI is reshaping each trade and predicted that each firm will both construct or lease AI factories to function their companies and help the intelligence of their merchandise. Via its world ecosystem of companions, NVIDIA helps companies transfer towards AI-native operations.
A number of trade leaders are already implementing these AI factories. Foxconn, Cadence, and Eli Lilly are among the many first to undertake the RTX PRO Server-powered infrastructure. Foxconn is utilizing the structure to develop AI capabilities throughout electronics manufacturing, sensible electrical automobiles, and robotics purposes. The corporate’s Chairman, Younger Liu, emphasised that NVIDIA Blackwell expertise will rework each facet of their world operations.
Prebuilt AI Software program Instruments
The RTX PRO 6000 Blackwell GPU permits a wide selection of enterprise workloads, from multimodal AI inference and scientific computing to video rendering and bodily simulation. NVIDIA RTX PRO Servers can help as much as eight of those GPUs and are geared up with BlueField-3 DPUs and ConnectX-8 SuperNICs for high-throughput, low-latency networking. This configuration additionally helps PCIe Gen 6 expertise for future-proof connectivity.
To maximise efficiency, the NVIDIA AI Enterprise platform provides prebuilt AI software program instruments akin to NVIDIA NeMo for generative fashions, NVIDIA NIM for optimized inference, and specialised AI Blueprints for creating digital people and AI question programs. These instruments are designed to work seamlessly with the {hardware} stack, permitting companies to deploy strong AI capabilities with minimal integration overhead.
The servers are additionally tightly built-in with NVIDIA Omniverse, a platform that permits simulation, industrial digital twin improvement, and robotics studying at scale. This makes RTX PRO Servers notably engaging to enterprises engaged on next-generation automation and digital engineering.
Reference Mannequin for Deploying AI Infrastructure
NVIDIA has additionally launched a validated reference structure for its Enterprise AI Manufacturing facility design. This offers enterprises with a confirmed mannequin for deploying AI infrastructure utilizing NVIDIA-Licensed Servers, networking {hardware}, and software program parts. The blueprint contains detailed steerage on software program stacks, deployment practices, and scaling methods to satisfy rising enterprise calls for.
To help large-scale AI duties, corporations also can use NVIDIA HGX B200 programs together with the AI Manufacturing facility design. These programs cater to organizations with superior AI workloads akin to coaching massive language fashions and operating real-time advice engines.
NVIDIA is collaborating with a broad group of system and expertise companions to carry these options to market. Main distributors akin to Cisco, Dell Applied sciences, Hewlett Packard Enterprise, and Lenovo will supply full-stack options utilizing RTX PRO Servers and NVIDIA AI Enterprise software. Further manufacturing companions embody ASUS, Supermicro, Pegatron, Wiwynn, and Quanta Cloud Expertise.
Storage infrastructure for these AI factories will probably be supplied by companions together with NetApp, IBM, Pure Storage, DDN, VAST Information, and others. These storage platforms are licensed to deal with the high-performance knowledge calls for of AI and accelerated computing.
On the companies facet, world consulting corporations akin to Accenture, Deloitte, EY, Infosys, Tata Consultancy Providers, and Wipro are serving to enterprise purchasers migrate to GPU-accelerated infrastructure utilizing the Enterprise AI Manufacturing facility mannequin. Their involvement ensures smoother deployments, strategic planning, and integration into broader digital transformation efforts.
Availability
The enterprise AI manufacturing facility with RTX PRO Servers may be deliberate by prospects by contacting their favourite NVIDIA expertise supplier. Uncover extra concerning the verified design of the Enterprise AI Manufacturing facility and start your journey on the NVIDIA Marketplace.
