Switch, a major AI data center operator, is partnering with SUSE and NVIDIA to create digital twin environments that simulate real-time power usage, thermal dynamics, and infrastructure performance. This approach allows data center managers to test changes in a virtual environment before implementing them in the physical world, reducing risk and downtime.

What Are Digital Twins and Why Do Data Centers Need Them?

A digital twin is a virtual replica of a physical system that continuously ingests operational data to model performance, predict outcomes, and optimize infrastructure. For data center operators managing massive AI facilities, digital twins solve a critical problem: the ability to simulate decisions around power distribution, cooling systems, and hardware configurations without requiring physical intervention.

The complexity of running modern AI data centers at scale makes this capability essential. Switch describes itself as the premier provider of AI, cloud, and enterprise data centers, and the company is using this digital twin initiative as part of its broader effort to manage and optimize infrastructure operations. The partnership was announced at SUSECON 2026, SUSE's annual customer and partner conference, held in Prague on April 22, 2026.

How Does the Switch-SUSE-NVIDIA Platform Work?

The technical architecture brings together three key components that traditionally operated separately. SUSE AI, built on SUSE Rancher Prime and SUSE Linux Enterprise Server, integrates with NVIDIA Omniverse simulation tools and DGX computing systems to deliver highly accurate digital twins of data center facilities.

A central innovation of this partnership is the elimination of siloed infrastructure that has historically separated high-end 3D graphics workloads from complex AI programs. Under the new architecture, language models, simulation, and real-time rendering can run on the same shared infrastructure simultaneously rather than across disconnected systems. This convergence happens on the NVIDIA DGX Platform, which serves as the shared hardware layer enabling physically accurate simulation alongside AI and machine learning processing.

• Unified Computing Layer: Language models, simulation, and rendering run on shared infrastructure instead of separate systems, maximizing hardware utilization
• Real-Time Simulation: The platform allows operators to model power usage, thermal dynamics, and infrastructure performance before changes are deployed physically
• Security and Reliability: The system functions in air-gapped environments, meaning it can remain secure without internet connectivity, and includes automated software update processes to reduce human error
• Enterprise AI Integration: The platform provides the necessary integration for large language models within a secure and manageable environment

Switch has framed this effort under its EVO AI Factory software systems and Living Data Center EVO platform, which it describes as the operating plane for unifying AI, simulation, and real-time operations.

What Problem Does This Solve for Data Center Operators?

Data center operators face a fundamental challenge: making infrastructure decisions that affect power consumption, cooling efficiency, and hardware performance without being able to test them safely first. Digital twins eliminate this constraint by allowing managers to run simulations before committing to physical changes. For an operator of Switch's scale, this capability could significantly impact decisions around power distribution, cooling systems, and hardware configurations.

"A new class of enterprise applications now requires language models, simulation, and rendering to converge within a single system rather than across disconnected silos. By integrating SUSE AI with NVIDIA DGX systems and Omniverse platforms, we enable these workloads to run on shared infrastructure, maximizing utilization while simplifying exascale operations," said Zia Syed, Chief Technology Officer at Switch.

Zia Syed, Chief Technology Officer at Switch

SUSE's contribution focuses on providing what the company calls a "digital floor" that ensures large AI workloads remain secure, manageable, and continuously available. SUSE AI is described as a fully governed, GPU-optimized enterprise AI platform designed to serve as the governed execution engine for deploying and orchestrating mission-critical AI applications across any infrastructure runtime.

"What we're enabling with Switch is the shift from experimentation to execution, where AI, simulation, and real-time rendering run side-by-side on the same infrastructure. By providing a resilient, open source foundation, SUSE gives leaders the flexibility to integrate best-in-class technologies, like NVIDIA AI Enterprise and accelerated computing, on their own terms," said Rhys Oxenham, General Manager of AI at SUSE.

Rhys Oxenham, General Manager of AI at SUSE

Why Does Open Source Matter in This Context?

SUSE positions its open source foundation as a key differentiator, giving customers the flexibility to integrate third-party technologies without being locked into proprietary systems. This approach is particularly important for large enterprises that want to avoid vendor lock-in while maintaining security and reliability standards.

Switch is already using the platform internally to run its own AI models, which the company says helps automate routine tasks and improve how it serves its clients. This real-world deployment demonstrates that the technology is moving beyond theoretical capability into practical operational use.

The partnership represents a significant shift in how data center operators approach infrastructure management. Rather than relying on physical testing or educated guesses, operators can now use digital twins to predict outcomes and optimize performance before changes are deployed. As AI data centers continue to grow in scale and complexity, this capability is likely to become increasingly essential for managing power consumption, cooling efficiency, and overall operational reliability.