NVIDIA’s 2024 GTC event, which runs until March 21, featured a wide range of announcements typical of a major tech conference. One announcement that stood out was the unveiling of the next-generation Blackwell GPU architecture during founder and CEO Jensen Huang’s keynote. This new architecture enables organizations to create and run real-time generative AI on trillion-parameter large language models.
“The future is generative…which is why this is a brand new industry,” Huang told attendees. “The way we compute is fundamentally different. We created a processor for the generative AI era.”
However, this was not the only ‘next-gen’ announcement made at the event in San Jose.
NVIDIA also presented a blueprint for building the next generation of data centers, promising “highly efficient AI infrastructure” with the support of partners such as Schneider Electric, Vertiv, and Ansys. The fully operational data center was showcased as a digital twin in NVIDIA Omniverse, a platform for creating 3D work tools, applications, and services. Additionally, cloud APIs were introduced to help developers seamlessly integrate core Omniverse technologies into existing design and automation software applications for digital twins.
The latest NVIDIA AI supercomputer is based on the NVIDIA GB200 NVL72 liquid-cooled system, featuring two racks with 18 NVIDIA Grace CPUs and 36 NVIDIA Blackwell GPUs connected by fourth-generation NVIDIA NVLink switches.
One of the partners mentioned in the announcement, Cadence, played a significant role with its Cadence Reality digital twin platform. This platform was described as the industry’s first comprehensive AI-driven digital twin solution for enhancing data center design and modernization, resulting in up to a 30% improvement in data center energy efficiency.
The demonstration showcased how digital twins can be used to test, optimize, and validate data center designs before physical production. Engineers were able to unify and visualize multiple CAD datasets with enhanced precision and realism, simulate airflows, and assess the performance of new liquid-cooling systems using Ansys software.
While the Blackwell GPU platform holds great promise, it requires a suitable infrastructure to operate, and major cloud providers are actively involved in offering the NVIDIA Grace Blackwell platform. AWS and NVIDIA announced that NVIDIA Blackwell on AWS will enable customers to unlock new generative artificial intelligence capabilities faster. Microsoft and NVIDIA are also collaborating to bring the GB200 Grace Blackwell processor to Azure.
In the healthcare and life sciences industries, AWS and Microsoft are working with NVIDIA to advance computer-aided drug discovery and improve efficiency in clinical research and care delivery. Google Cloud is integrating NVIDIA NIM microservices into its platform to expedite generative AI deployment, and NexGen Cloud is offering sustainable infrastructure as a service with its Hyperstack platform powered by renewable energy.
NexGen Cloud announced that NVIDIA Blackwell platform-powered compute services will be part of the AI supercloud, providing customers with powerful GPU offerings for driving innovation and achieving unprecedented efficiencies.
