“I hope you realize this is not a concert,” said Nvidia president Jensen Huang to a packed SAP Center in San Jose. CEO Jensen Huang delivered an electrifying keynote at NVIDIA’s GPU Technology Conference (GTC) 2024, making groundbreaking announcements promising to redefine technology. With over 11,000 attendees and a global audience, the event excited and speculated about the future of accelerated computing and AI.
Table of contents
Glimpse of Crowd at GTC 2024
You have arrived at a developers conference. There will be a lot of science describing algorithms, computer architecture, mathematics. I sense a very heavy weight in the room; all of a sudden, you’re in the wrong place.
Jensen Huang CEO of NVIDIA
Let’s look at the highlights of NVIDIA GTC 2024 Keynote!
Accelerated Computing
Huang highlighted the accelerated computing paradigm as essential for sustaining the growth of computational demands across industries. With general-purpose computing reaching its limits, NVIDIA’s accelerated computing architecture promises substantial performance improvements, fostering advancements in fields ranging from climate science to healthcare. The shift towards accelerated computing is not just technological but marks a new industrial revolution, demanding modernization of the global data center infrastructure.
Generative AI: Creating a New Category of Software
The keynote spotlighted generative AI, emphasizing its emergence as a poweful technology capable of creating software that never existed before. This new category of AI, exemplified by advancements like ChatGPT and various generative models, showcases the ability to generate novel content, including text, images, and even software code, heralding a new era of AI-driven creativity and productivity.
Introduction to Blackwell Architecture
At the heart of Huang’s keynote was the introduction of the Blackwell architecture, a revolutionary leap forward named after David Harold Blackwell, the esteemed mathematician. It significantly surpasses the capabilities of its predecessor, Hopper, challenging traditional computing paradigms.
Technical Innovations
- Transistor and Process Innovation: Blackwell is engineered with 208 billion transistors across two chips, utilizing a sophisticated 4NP TSMC process. This design enables a remarkable 10TB/s interface, pushing beyond conventional semiconductor limitations.
- Efficiency and Power: The architecture introduces FP4 and FP6 compute capabilities, enhancing model training efficiency with a balance of power and precision. A second-generation transformer engine drives substantial improvements in compute, bandwidth, and model size, catering to AI development needs.
How is Blackwell going to help? Let’s explore that in the section below:
Bandwidth and Reliability
- Advanced Connectivity: Integration with PCIe Gen6 and HBM3e memory technology significantly increases bandwidth. The fifth-generation NVLink doubles bandwidth to 1.8TB/s, supported by the RAS Engine for improved AI deployment reliability.
- Confidential Computing: Blackwell also introduces capabilities for confidential computing, ensuring secure data processing within trusted execution environments on GPUs.
Source: TechRepublic
You can also this latest Video of NVIDIA CEO Jensen Huang Shows the Blackwell
Blackwell’s Technological Advancements
- Performance Milestones: Achieving 2.5x performance in FP8 for training and 5x in FP4 for inference, the Blackwell chip, alongside the NVIDIA GB200 Grace Blackwell Superchip, sets new standards for GPU performance and memory architecture efficiency.
- Networking Platforms: The integration with Quantum-X800 InfiniBand and Spectrum-X800 Ethernet platforms enables networking speeds up to 800Gb/s, while the NVLink Switch chip aids in constructing the advanced NVIDIA GB200 NVL72 system for AI training and inference.
- Supercomputing Capabilities: The DGX SuperPOD, powered by NVIDIA GB200 Grace Blackwell Superchips, is a landmark in AI supercomputing. Designed for trillion-parameter models, it provides 11.5 exaflops of computing power at FP4 precision, featuring a liquid-cooled architecture for scalable, high-efficiency AI processing.
Industry Adoption and Support
- Corporate Support: Prominent figures like Sundar Pichai (Alphabet and Google), Elon Musk (Tesla and xAI), and other industry leaders have publicly endorsed Blackwell, indicating a widespread industry shift towards NVIDIA’s vision.
- Global Adoption: With the backing of leading cloud services and innovative AI companies, Blackwell’s comprehensive adoption showcases its potential to revolutionize various sectors.
Generative AI with NIM
Huang’s vision extended beyond hardware into how we create and deploy software. He introduced NVIDIA NIM (NVIDIA inference microservices), a game-changer in packaging and delivering AI-driven software. This innovation connects developers with millions of GPUs worldwide, allowing for the deployment of custom AI across various applications.
Merging AI with the Real World: Omniverse Cloud APIs
- Omniverse Cloud Integration: Highlighting the fusion of AI with physical reality, Omniverse Cloud APIs deliver advanced simulation capabilities essential for the development of robotics and AI’s understanding of our world.
- A Paradigm Shift: Generative AI is redefining software development processes. Instead of traditional coding, companies will leverage AI models, assign them tasks, and guide them with examples and feedback. NVIDIA NIMs, built on NVIDIA’s computing libraries and generative AI models, represent this new approach.
- Simplifying Integration: NVIDIA’s microservices are designed to be easily integrated into existing systems, supporting standard industry APIs. They are optimized for NVIDIA’s CUDA platform, continuously updated for new GPU generations, and regularly scanned for security to ensure robust and safe AI solutions.
Project Groot: The Future of Humanoid Robotics
Among the myriad of announcements, Project Groot emerged as a beacon of NVIDIA’s ambition in humanoid robotics. Named playfully yet with a nod to the popular “Guardians of the Galaxy” character, Groot represents NVIDIA’s venture into general-purpose foundation models for humanoid robots. Huang envisioned a world where robotics and AI converge to create autonomous systems capable of interacting seamlessly with their environment. This project aims to bridge the gap between AI’s digital prowess and its physical manifestations, learning from human examples and adapting to the laws of physics through rigorous training in virtual environments like Omniverse.
The Foundation: Jetson Thor and Isaac Lab
The Jetson Thor computer and Isaac Lab form the core of Project Groot’s development infrastructure. Jetson Thor, equipped for Transformer engines, is set to provide unmatched processing power for humanoid robots, enhancing their ability to navigate and interact with the world with unprecedented precision.
The Future of Robotics
Huang predicts a future where all moving things will incorporate robotics, with the automotive sector playing a significant role. Highlighting NVIDIA’s influence, Huang shared that BYD, the leading electric vehicle manufacturer, has chosen NVIDIA’s next-gen computers for its autonomous vehicles.
Advancements in Perception and Manipulation
- Isaac Perceptor SDK: Introduced to enhance robots’ environmental perception, this toolkit offers advanced visual odometry, 3D reconstruction, and depth sensing capabilities.
- Isaac Manipulator: This new library advances robotic arm technology with sophisticated perception, path planning, and control mechanisms, signaling major progress in robotic adaptability.
Project GR00T, distinct from Groot, represents NVIDIA’s commitment to advancing robotics and embodied AI. Alongside the announcement of Project GR00T, NVIDIA introduced the Jetson Thor computer tailored for humanoid robots and substantial enhancements to the NVIDIA Isaac robotics platform, laying the groundwork for more sophisticated robotic systems.
GTC March 2024 Keynote with NVIDIA CEO Jensen Huang
End Note
NVIDIA’s GTC 2024 keynote, led by Jensen Huang, was more than a presentation; it was a declaration of the future. With Blackwell architecture revealed and Project Groot in sight, NVIDIA is shaping a tech era set to transform our digital and physical realms. Standing on this cusp, NVIDIA’s innovations might ignite the next industrial revolution, driving progress through accelerated computing and generative AI.
Stay tuned to Analytics Vidhya Blogs to know more about what’s happening in the world of GenAI!
