Is this page helpful?

Overview#

The NVIDIA Omniverse DSX Blueprint for AI Factory Digital Twins is built for developers to accelerate the design through operation of gigawatt-scale AI factories by integrating physical and digital data into interactive digital twins built on OpenUSD. By leveraging SimReady assets and Omniverse acceleration libraries, developers can incorporate real-time power, thermal, and operational simulations directly into their own workflows to improve efficiency, sustainability, and reliability.

The blueprint provides an end-to-end framework to build, simulate, and optimize gigawatt-scale AI data centers — seamlessly transitioning from custom app development and physically accurate design to high-efficiency, grid-resilient operations.

What’s Included#

The blueprint repository contains:

  • Digital Twin geometry based on the entire DSX reference design for a 50-acre site including compute building and support infrastructure.

  • Front-end web application with a user interface developed with Omniverse libraries for interacting with digital twins, viewing simulations, and creating and saving build configurations.

  • Simulation-ready assets to accelerate digital twin creation:

    • Computational Fluid Dynamic (CFD) thermal hot-aisle simulation

    • Sample compute configurations for DSX such as GB200 and GB300 NVL72 designs

    • Electrical loading simulation to test various loading configurations

The blueprint is powered by NVIDIA libraries including:

Key Capabilities#

  • Accelerated Time-to-Revenue: Streamline operations through system-level optimization, modular construction, and rapid cluster bring-up.

  • Energy-Efficient Performance: Maximize token throughput per power budget while securing early access to power resources.

  • Sustainable Resource Management: Minimize environmental impact using dry coolers, heat reuse, and advanced water-usage optimization.

  • Intelligent Workload Placement: Coordinate IT and OT systems to automate workload distribution based on real-time power and cooling availability.

  • AI-Driven Reliability: Ensure maximum uptime via integrated AI agents and predictive maintenance across IT and infrastructure layers.

  • Future-Proof Infrastructure: Design flexible environments capable of supporting multiple hardware generations with minimal downtime.

User Roles#

The blueprint supports a range of disciplines across the AI factory lifecycle:

  • Design Engineers author detailed 3D geometry and configuration in DCC/CAD applications, publishing assets into the USD Simulation Schema as the canonical design baseline.

  • Network Administrators use NVIDIA Air to model and validate network topology, contributing connectivity and bandwidth constraints into the shared USD models.

  • Mechanical Engineers use Cadence simulation tools to generate thermal and structural results linked to the corresponding USD entities.

  • Electrical Engineers use ETAP to produce electrical simulation data — load flows, fault studies, and protection settings — associated with the shared USD models.

  • Reviewers interact with hosted applications through Omniverse App Streaming, accessing real-time 3D visualizations and simulation results without local high-end hardware.

  • AI Engineers utilize trained foundation models through AI Training ISV applications to compose custom simulation and inference workflows.

For a detailed view of how data flows between these roles and the system components, see System Architecture.

In this guide, you learn how to:

  • Set up your development environment and install dependencies

  • Build and run the Kit application

  • Connect the web frontend to the streaming server

  • Navigate the user interface and interact with 3D scenes

  • Understand the system architecture and communication flow

Important

This blueprint is designed to provide an example of integrating the workflow for developers and demonstrate key concepts and patterns. It is not a turn-key application ready for production deployment without customization.

Developers are expected to use this guide as a starting point and extend the blueprint according to their specific requirements, potentially making significant architectural or implementation changes as needed for their particular use cases.

Getting Started#

This guide focuses on local development. Follow these steps to get started:

  1. Review Prerequisites — Verify system requirements and install dependencies

  2. Follow Quick Start — Clone repository, build, and run locally (15-20 minutes)

  3. Explore User Interface Walkthrough — Learn the web portal features

  4. Understand DSX Applications — Learn about different Kit application configurations

  5. Review System Architecture — Understand how components work together

Cloud Deployment#

If you want to deploy DSX to cloud infrastructure, see the Application Streaming section for container deployment and streaming configuration.