Anirudh Devgan and Jensen Huang - CadenceLIVE Silicon Valley 2024 - Fireside Chat

Video Description

Hear from Jensen Huang at #CadenceLIVE as he discusses the pivotal role of AI and accelerated computing in shaping industry mega-trends and how #Nvidia and #Cadence are collaborating to drive transformational change across EDA, SDA, digital biology and AI. We are witnessing a fundamental transition in the history of electronic design automation. As Jensen Huang points out, traditional CPU scaling has hit a wall. To meet the demands of modern ai, the industry must move toward accelerated computing. This shift isn't just about faster chips; it’s about a 1,000x factor in compute capability that serves as the foundation for the next generation of generative ai. Without this transition, the energy and cost requirements of the modern data center would become unsustainable. The creation of the NVIDIA Blackwell GPU serves as the perfect case study for the co-design philosophy. 1. The Digital Twin of Silicon: Jensen describes Palladium not just as a tool, but as the "single most important appliance" in his life. Before silicon even exists, the entire system is simulated to ensure functional safety and performance. 2. Full-Stack Optimization: By treating the data center as the new unit of computing, NVIDIA and Cadence can innovate across the entire stack—from the graphics processing unit to the liquid-cooled racks. This allows for hyperscale efficiency where 3% of the code can be accelerated to provide 99% of the performance gains. 3. The AI Jump: Jensen notes that we can no longer explore the infinite multidimensional design space through brute force. We need ai to "jump" into high-potential areas of optimization, followed by principled solvers to home in on the best answer. The partnership extends beyond silicon into the worlds of digital biology and robotics such as : 1. Biology as Engineering: By applying simulation and eda principles to biology, the industry is turning "drug discovery" into "drug engineering." 2. Humanoid Robotics: The next generation of autonomous machines will rely on the same fail-safe operating systems used in automotive, potentially bringing the cost of a humanoid robot down to $10,000–$20,000. 3. Sustainability through Density: Despite high power consumption, accelerated computing is significantly more energy-efficient per unit of work done. Moving from tens of thousands of general-purpose servers to a single dense ai cluster reduces energy waste by 20x or more. How did NVIDIA transform from a chip company to a full-stack system company without losing its agility? Watch the full video to hear Jensen’s 'Gladiator' management philosophy and learn why he believes the Blackwell GPU would have been impossible to build without a trillion-simulation pre-verification flow. Video Highlights & Timestamps : 0:04 – Anirudh Devgan Welcomes Jensen Huang to CadenceLIVE 2024 0:48 – Technical Partnership: Why Blackwell Required Palladium Simulation 2:55 – The AI Evolution: Moving from CPU Scaling to Accelerated Computing 4:32 – Exploring the Multidimensional Design Space with Artificial Intelligence 6:37 – Full-Stack Innovation: Transforming from EDA to Co-design (SDA) 7:53 – Energy Efficiency: Saving Power for a Billion People through Better Design 10:58 – Domain-Specific Acceleration: Why One Processor Cannot Run All Code 13:21 – Digital Biology: Turning Drug Discovery into an Engineering Field 15:58 – The Data Center as a Supercomputer: Autonomous Systems and Safety 17:40 – The Future of Robotics: Why Humanoid Robots are the Next Frontier 21:12 – Managing Power Density in High-Performance AI Infrastructure 25:13 – Jensen Huang’s Management Style: Transparency and Empowerment 27:53 – Conclusion: The Remarkable Transformation of the Computing Industry ------------------------------------------------------------------------------------------------------------------------------------------------------------- Connect with Cadence : • YouTube: https://www.youtube.com/@cadencedesignsystems • LinkedIn: https://www.linkedin.com/company/cadence/ • Facebook: https://www.facebook.com/CadenceDesign • Twitter/X: https://twitter.com/Cadence • Instagram: https://www.instagram.com/cadencedesignsystems/ ------------------------------------------------------------------------------------------------------------------------------------------------------------- About Cadence : Cadence is a global leader in electronic systems design, applying its Intelligent System Design™ strategy to deliver software, hardware, and IP that turn design concepts into reality. Cadence® customers are among the world’s most innovative companies, creating breakthrough products — from chips to boards to full systems — across markets including hyperscale computing, 5G, automotive, aerospace, mobile, consumer, industrial, and healthcare. Recognized by Fortune as one of the 100 Best Companies to Work For ten (10) consecutive years. Learn more at https://www.cadence.com.