Dell and NVIDIA Introduce New Solutions for AI Data Centers and Enterprise Infrastructure

Dell Technologies and NVIDIA are expanding their strategy for building AI infrastructure, placing data centers at the core of enterprise AI adoption. The new developments are part of the Dell AI Factory with NVIDIA initiative, introduced in March 2024 and already used in over 4,000 customer deployments.

The platform combines GPU acceleration, networking, and software solutions to support workloads such as retrieval-augmented generation (RAG) and multimodal search, which processes both text and images.

Focus on AI Infrastructure in Data Centers

At the core of the new solutions are PowerEdge servers, which are essential for handling AI workloads – both for model training and inference (real-time operations).

Among the new systems are:

• PowerEdge XE9812, based on the NVIDIA Vera Rubin NVL72 platform for high-performance AI tasks
• XE9880L, XE9882L, and XE9885L, using the NVIDIA HGX Rubin NVL8 architecture and equipped with liquid cooling for more efficient thermal management
  
  

Liquid cooling is becoming increasingly important for modern data centers, as it enables dense GPU configurations without increasing power consumption.

Additionally, Dell introduced the PowerEdge R770, R7715, and R7725 servers, which enable the integration of AI capabilities into existing infrastructure through NVIDIA RTX PRO 4500 Blackwell Server Edition GPUs.

Networking for AI Workloads

As AI workloads increase, network infrastructure becomes a critical factor. Dell’s new solutions include:

• PowerSwitch SN6000 series, offering speeds up to 1.6 Tbps and support for liquid cooling
• PowerSwitch SN5610 and SN2201, supporting multiple operating systems such as Cumulus Linux and Enterprise SONiC
  
  

The ecosystem is further enhanced with NVIDIA Quantum-X800 InfiniBand Q3300-LD switches, which provide low latency and high throughput – key for AI and cloud environments.

These solutions ensure that data processing is not limited by network bottlenecks.

From Development to Production

Dell is also extending AI capabilities to developers through workstation solutions such as Dell Pro Max with GB10 and GB300, based on the NVIDIA GB300 Grace Blackwell chip. This enables local development and testing of AI models before deploying them into production environments.

This approach simplifies the transition from experimentation to real-world deployment – a process that remains challenging for many organizations.

Integrated Systems for Scalability

To simplify deployment, Dell also offers Integrated Rack Scalable Systems, which combine servers, networking, and infrastructure into ready-to-use configurations. This reduces complexity in building AI environments and accelerates deployment in data centers.

Direction: From AI Pilot to Scale

The new solutions, presented at NVIDIA GTC 2026, demonstrate Dell and NVIDIA’s effort to unify compute power, networking, and infrastructure into a single ecosystem.

The goal is to help organizations transition from experimental AI projects to full-scale production environments, where AI becomes a real driver of business.