Announcing Dell PowerEdge XE9680L Featuring AI Factory Rack-Scale Architecture

The GPU-accelerated Dell PowerEdge XE9680 is one of Dell’s fastest growing products since the PowerEdge product line was introduced 30 years ago. Hot on the heels of the Dell PowerEdge.Next announcement for Cloud Service Providers and near-edge deployments, Dell unveiled an exciting new addition to the PowerEdge XE product family. The Direct Liquid Cooled (DLC) Dell PowerEdge XE9680L with NVIDIA Blackwell Tensor Core GPUs represents a significant leap forward in AI computing. As part of the Dell AI Factory with NVIDIA announcements at Dell Technologies World 2024, this server redefines the boundaries of performance, GPU density per rack and scalability for AI workloads.

At the heart of the XE9680L lies its smart cooling and innovative rack-scale architecture. Here’s why it matters:

GPU Density per Rack, Efficient Cooling and Leading Performance

• • The XE9680L is designed for the most demanding large language model (LLM) training and large-scale AI inferencing environments, where GPU density per rack is critical. With a compact 4U form factor, it delivers one of the industry’s highest density x86 server solutions for the next-gen NVIDIA HGX B200.
  • The XE9680L leverages efficient DLC smart cooling for both CPUs and GPUs. This breakthrough technology allows for a more rack-dense 4U configuration, maximizing compute power while maintaining thermal efficiency.
  • The XE9680L is optimized for the next-gen NVIDIA HGX B200, providing exceptional performance for training large language models (LLMs) and other AI workloads.

More PCIe 5 Expansion Capability

• • The XE9680L boasts standard 12 x PCIe 5.0 full-height, half-length slots, giving customers 20% more FHHL PCIe 5.0 density.¹ This means 2x more² capacity for high-speed I/O for North/South AI fabric, Dell PowerScale GPU Direct storage connectivity and seamless integration of accelerators.
  • Whether you’re deploying deep learning models, running simulations or handling data-intensive tasks, the XE9680L’s PCIe capacity ensures smooth data flow.
    

Turn-key Solution with Rack-scale Factory Integration

• • Dell’s commitment to quality extends to the entire lifecycle of the XE9680L. With rack-scale factory integration, partner components are seamlessly integrated, ensuring a reliable and efficient deployment process.
  • Say goodbye to deployment headaches and hello to accelerated AI workloads delivered with the fastest time-to-value. From rack, stack and cabling to PDU sizing, the XE9680L is a turn-key solution.
  • Scale up to 72 Blackwell GPUs per 52 RU rack or 64 GPUs per 48 RU rack with the Dell PowerEdge XE9680L.
  • Take the guesswork out of scaling power, cooling and AI fabric with pre-validated rack infrastructure solutions.
  • Factory-integrated rack-scale AI factory solutions delivered to your data center or colocation facility floor with “one call” support and expert deployment services.

Ready to explore the future of AI computing? Visit Dell.com/XE to chat with a Dell Business Advisor. Discover how the XE9680L can transform your AI infrastructure, boost productivity and drive innovation. The Dell PowerEdge XE9680L will be available globally in the second half of this year.

Disclaimer: The information provided in this blog post is based on publicly available data as of the time of writing. Actual specifications and features may vary. Please refer to official Dell documentation for the most up-to-date details.

¹ Compared to Supermicro SYS-421GE-TNHR2-LCC.
² Based on Dell configuration guidance for 8 PCIe Gen 5 FHHL slots 1:1 allocated to the 8 GPUs in the XE9680 and XE9680L. The XE9680 has 10 PCIe Gen 5 FHHL slots while the XE9680L has 12 PCIe Gen 5 FHHL slots.