AI & Simulation: Reshaping Future Tech 🚀💡

CADENCE EXPANDS AI PARTNERSHIPS WITH NVIDIA AND GOOGLE CLOUD
Cadence Design Systems unveiled two significant AI collaborations at its CadenceLIVE event, solidifying partnerships with Nvidia and Google Cloud. These initiatives aim to revolutionize system-level design and robotics through the fusion of physics-based simulation and accelerated computing.

COLLABORATION WITH NVIDIA: SIMULATION AND ACCELERATED COMPUTING FOR ROBOTICS
The core of Cadence’s Nvidia partnership centers around leveraging AI to enhance physics-based simulation and accelerated computing, specifically tailored for robotic systems and large-scale AI infrastructure. This approach, termed “physical AI” by Nvidia, focuses on modeling and deploying solutions within semiconductors and AI infrastructure. Cadence is integrating its multi-physics simulation and system design tools with Nvidia’s CUDA-X libraries, AI models, and Omniverse-based simulation environment. These tools are designed to accurately model thermal and mechanical interactions, allowing engineers to assess system behavior under realistic operational conditions. The integration extends beyond traditional chip design to encompass infrastructure components such as networking and power systems, creating a holistic platform for simulating system performance before physical deployment. The collaboration explicitly recognizes the interconnectedness of compute, networking, and power systems in determining overall system performance. Nvidia CEO Jensen Huang emphasized this collaboration, stating that the companies are working together on robotic systems.

ROBOTICS DEVELOPMENT AND SIMULATION DATA GENERATION
Cadence’s physics engines, which model the interaction of real-world materials, are being integrated with Nvidia’s AI models used to train AI-driven robotic systems within simulated environments. This strategy significantly reduces the reliance on costly and time-consuming real-world data collection. The companies highlighted that training datasets should be generated using physics-based models rather than collected from physical systems. The accuracy of these generated datasets directly impacts the performance of the AI models. Companies like ABB Robotics, FANUC, YASKAWA, and KUKA are integrating these simulation tools into virtual commissioning workflows, enabling them to test production systems in software prior to physical rollout. Nvidia’s systems model complex robot operations and entire production lines using physically accurate digital environments.

CHIP DESIGN AUTOMATION ON GOOGLE CLOUD: AI-POWERED LAYOUT PROCESSES
Cadence introduced a new AI agent designed to automate later-stage chip design tasks, specifically focusing on physical layout processes. This agent translates circuit designs into silicon implementations, building upon an earlier agent introduced for front-end chip design, which handles circuit design. The new agent utilizes Google’s Gemini models for automated design and verification workflows, leveraging the power of the cloud. The system, known as the ChipStack AI Super Agent platform, employs model-based reasoning with native design tools to coordinate tasks across multiple design stages, interpreting design requirements and automatically executing tasks. Early deployments have reportedly yielded productivity gains of up to 10 times in design and verification tasks. Cadence emphasized their role in building AI systems, with those systems then assisting in optimizing the design process.

DIGITAL TWINS AND SYSTEM VALIDATION
Digital twin models are employed to validate systems in virtual environments before physical deployment. These models allow engineers to test design trade-offs, evaluate performance scenarios, and optimize configurations – all within a software-based simulation. The cost and complexity associated with large-scale data center infrastructure traditionally limited the use of trial-and-error deployment methods, but digital twins provide a cost-effective alternative.

QUANTUM AI MODELS: NVIDIA ISING FOR QUANTUM PROCESSOR CALIBRATION
Nvidia concurrently announced a family of open-source quantum AI models, dubbed NVIDIA Ising. These models are named after the Ising model, a mathematical framework used to represent interactions in physical systems. The Ising models are specifically designed to support quantum processor calibration and quantum error correction. Nvidia reported that the models achieve up to 2.5 times faster performance and three times higher accuracy in decoding processes used for error correction. Huang stated that “AI is essential to making quantum computing practical,” and that the Ising models transform fragile qubits into scalable and reliable quantum-GPU systems, positioning AI as the “operating system” of quantum machines.

AI-POWERED CONTROL PLANE FOR QUANTUM MACHINES
The NVIDIA Ising models are intended to function

INDUSTRY EVENTS AND COLLABORATION OPPORTUNITIES
To further explore these advancements and connect with industry leaders, Cadence and Nvidia co-hosted the AI & Big Data Expo in Amsterdam, California, and London. This event, part of TechEx and co-located with other leading technology events, provided a platform for discussions around AI and big data solutions. TechForge Media powers AI News, offering ongoing coverage and insights into the evolving landscape of enterprise technology.