NVIDIA Rethinks AI: Next Gen Data Centers, Supercomputers For Healthcare & More at GTC Day 1

“If the last 20 years were amazing, the next 20 will seem nothing short of science fiction.”

Jensen Huang, CEO, NVIDIA

NVIDIA’s CEO, Jensen Huang, kicked off the GPU Technology Conference in style with his kitchen keynote underlining the various ways in which his team has pushed the envelope of AI as a domain. This five-day conference flaunts 500+ talks from industry experts along with hot announcements from NVIDIA. In this article, we bring you the highlights from the day 1 of GTC.

NVIDIA By The Numbers


Sign up for your weekly dose of what's up in emerging technology.
  • NVIDIA has shipped over 1 billion CUDA compatible GPUs.
  • 6,500 startups are building on NVIDIA.
  • CUDA SDK has been downloaded 20 million times, and 6 million times this year alone.
  • 1,800 applications are now CUDA-accelerated.
  • There are over 2 million NVIDIA developers now.

NVIDIA For Next Gen Data Centers

Data centre’s infrastructure can consume up to 30% of its CPU cores. And as traffic within a data centre and microservices increase, this load will increase dramatically.

NVIDIA believes that the next-generation data centres require a new kind of processor. They have built DPUs for this purpose. These DPUs consist of accelerators for networking, storage, security and programmable Arm CPUs to offload the hypervisor. As the latest addition to the DPU family, Huang announced BlueField 2 DPU.

Powered by Mellanox technology, the new NVIDIA BlueField 2 DPU is a programmable processor with powerful Arm cores and acceleration engines. Huang revealed that the BlueField-2 is sampling now, BlueField-3 is finishing, and BlueField-4 is in high gear.

“We are going to bring a ton of technology to networking. In just a couple of years, we’ll span nearly 1,000 times in compute throughput,” said Huang, talking about the DPU.

Along with a new DPU, NVIDIA also announced DOCA, its programmable data-centre-infrastructure-on-a-chip architecture, which enables developers to write infrastructure apps for software-defined networking, cybersecurity, telemetry and more.

All Things ARM

CEO Jensen with Ampere GPU

As expected day one consisted of key announcements and panel discussions concerned with ARM. NVIDIA pocketed ARM for a staggering $40 billion and have teased their ambitious objectives last month. “Arm is the most popular CPU in the world,” Huang said. “Together, we will offer NVIDIA accelerated and AI computing technologies to the Arm ecosystem.”

Yesterday, Huang announced a major initiative to advance the Arm platform:

  • NVIDIA and Arm partners will work together to create platforms for HPC, cloud, edge and PC.
  • It will also support Arm partners with GPU, networking, storage and security technologies to create complete accelerated platforms.
  •  NVIDIA AI and NVIDIA RTX engines will be ported to Arm.

By combining in house technologies, NVIDIA wants to make Arm platforms a leader in edge at accelerated and AI computing.

A Supercomputer For Healthcare

Huang spoke about NVIDIA’s effort to build the U.K.’s fastest supercomputer — Cambridge-1. With this, NVIDIA plans to bring state-of-the-art computing infrastructure to “an epicentre of healthcare research.” On completion, Cambridge-1 will boast 400 petaflops of AI performance, which shall place it in the upper echelons of supercomputers. Cambridge-1 will host NVIDIA’s U.K. AI and healthcare collaborations with academia, industry and startups.

NVIDIA’s ambitious healthcare pursuits are backed up by pharma biggies like AstraZeneca, GSK; King’s College London; the Guy’s and St Thomas’ NHS Foundation Trust and startup Oxford Nanopore.

AI At The Edge

NVIDIA has invested heavily in building integrated software and hardware solutions for the edge. At GTC, the company has announced few of their developments that will democratise AI like never before.

  • NVIDIA’s EGX platform is expanding to combine the Ampere architecture GPU and BlueField-2 DPU on a single PCIe card. The NVIDIA EGX AI platform is built to set up a state-of-the-art edge-AI server and can be deployed at factories, perform automatic checkout at retail or help nurses monitor patients.
  • NVIDIA Fleet Command, a new service was announced using which edge computing can be leveraged across IoT devices, which combines the security and real-time processing with the remote management and ease of software-as-a-service.
  • In an aggressive push to democratise robotics, NVIDIA is launching Jetson Nano 2GB, the latest addition to the Jetson family. Jetson is an Arm-based System on Chip designed for robotics. Thanks to the sensor processors, the CUDA GPU and Tensor Cores, and, most importantly, the richness of AI software that runs on it. According to Huang, AI software is a big breakthrough that will make robots smarter and more adaptable. But it’s the NVIDIA Jetson AI computer that will democratise robotics.

Building Chatbots And Recommenders Gets Better

NVIDIA is bringing its much talked about services Jarvis and Merlin for all. Huang announced that NVIDIA Jarvis for conversational AI services and NVIDIA Merlin for recommender systems had entered open beta. This will now allow companies to explore larger deep learning models and develop more nuanced and intelligent systems. Businesses can leverage Conversational AI services built on Jarvis for building chatbots and recommender systems built on Merlin for eCommerce and other applications.

For instance, a chatbot that serves in real-time should be able to mount model computations in under 300 milliseconds.

Jarvis ability to handle multiple data streams in real-time enables the delivery of improved services. It enables more natural interactions through sensor fusion — the integration of video cameras and microphones. 

Stay tuned to Analytics India Magazine for more updates from GTC Day 2

More Great AIM Stories

Ram Sagar
I have a master's degree in Robotics and I write about machine learning advancements.

Our Upcoming Events

Conference, in-person (Bangalore)
Machine Learning Developers Summit (MLDS) 2023
19-20th Jan, 2023

Conference, in-person (Bangalore)
Rising 2023 | Women in Tech Conference
16-17th Mar, 2023

Conference, in-person (Bangalore)
Data Engineering Summit (DES) 2023
27-28th Apr, 2023

Conference, in-person (Bangalore)
MachineCon 2023
23rd Jun, 2023

3 Ways to Join our Community

Discord Server

Stay Connected with a larger ecosystem of data science and ML Professionals

Telegram Channel

Discover special offers, top stories, upcoming events, and more.

Subscribe to our newsletter

Get the latest updates from AIM