Korean
Dell Technologies has launched the PowerEdge XE8812, a new liquid-cooled server platform optimized for large-scale AI inference and high-performance computing (HPC) workloads. Integrated into the Dell AI Factory with NVIDIA portfolio, the system is built on the NVIDIA Vera Rubin NVL4 architecture, supporting up to 144 densely configured GPUs per rack.
This launch addresses the rising demand for advanced infrastructure to power increasingly complex AI and simulation workloads. Scientific research, engineering, national AI programs, and genomics sectors now require higher compute density, memory capacity, and power efficiency than prior-generation systems can deliver.
Built for Converged AI and HPC Workloads
The fanless, direct liquid-cooled PowerEdge XE8812 is designed for large-scale simulation and data-heavy workloads, covering use cases including molecular modeling, multiphysics simulation, and foundation model development.
Compared with NVIDIA GB200 NVL4-based systems, the new Vera Rubin NVL4 platform boosts host memory, CPU core count, GPU memory, and overall compute resources. Dell upgrades CPU cores from 144 to 176, with 50% more per-socket system memory and GPU memory than the previous generation. This upgrade enables larger AI models and simulations to run fully in resident memory.
In-memory model and dataset residency reduces reliance on latency-prone data staging and swapping operations, which introduce microsecond-to-millisecond delays and lower effective bandwidth. Such overhead severely hinders performance in HPC and AI environments, where data movement often dominates overall application throughput. The platform also leverages NVIDIA CUDA-X libraries to accelerate scientific and AI workloads, enabling larger simulations and models without extra infrastructure expansion.
High-Density Rack-Scale Design with Liquid Cooling
Dell will deliver the XE8812 on an Open Rack v3 (ORv3) architecture, supporting 144 GPUs per rack and over 300kW rack power delivery. Direct liquid cooling for both CPUs and GPUs enables higher component density and superior thermal efficiency.
The modular ORv3 design simplifies deployment and integration within large-scale data centers. Dell’s integrated management suite further streamlines daily operations. The Integrated Dell Remote Access Controller (iDRAC) supports remote deployment, monitoring, and updates for PowerEdge servers. At the rack level, Dell Integrated Rack Controller and OpenManage Enterprise deliver real-time telemetry, system monitoring, and automatic leak detection, enhancing operational visibility and mitigating infrastructure risks.
Factory-Integrated Rack Deployment
Dell pairs the XE8812 with Dell PowerRack, a factory-integrated rack-scale solution that lowers deployment complexity for large AI and HPC clusters. Pre-validated PowerRack configurations shorten the transition from installation to production. Combined with Dell ProDeploy professional services, the pre-integrated racks eliminate manual integration work and can run live workloads in just over six hours.
Expanding Dell AI Factory Global Deployment
Dell’s AI Factory initiative continues rapid growth, with over 5,000 global customers adopting its AI infrastructure solutions. Major deployments across regions demonstrate its versatile workload support capabilities.
In the U.S., Dell, NVIDIA, and the National Energy Research Scientific Computing Center (NERSC) are building the Doudna supercomputer at Lawrence Berkeley National Laboratory. Equipped with XE8812 servers, NVIDIA Vera Rubin NVL4 technology, and NVIDIA Quantum-X800 InfiniBand networking, the system supports large-scale scientific computing and data-intensive research.
In France, AI firm InstaDeep is scaling its Kyber supercomputing cluster with Dell AI Factory infrastructure. Delivering around 0.5 exaFLOPs of FP16 performance, the cluster supports industrial design scenarios such as automated printed circuit board development.
In the UK, the Wellcome Sanger Institute leverages Dell PowerEdge XE-Series servers and NVIDIA GPUs for genomic research. The institute now assembles a complete genome every seven hours and manages over 100 PB of in-house curated genomic data, contributing more than 70% of genomes to the global Earth BioGenome Project.
In Australia, Monash University has deployed the MAVERIC supercomputer in partnership with Dell, NVIDIA, and CDC Data Centres. Powered by liquid-cooled Dell PowerRack infrastructure and GB200 NVL72-based PowerEdge XE9712 servers, it supports AI and data-driven research in cancer detection, climate action, and genomics.
Availability
Dell will launch the PowerEdge XE8812 globally in early 2027.
Beijing Qianxing Jietong Technology Co., Ltd.
Sandy Yang/Global Strategy Director
WhatsApp / WeChat: +86 13426366826
Email: yangyd@qianxingdata.com
Website: www.qianxingdata.com/www.storagesserver.com
Business Focus:
ICT Product Distribution/System Integration & Services/Infrastructure Solutions
With 20+ years of IT distribution experience, we partner with leading global brands to deliver reliable products and professional services.
“Using Technology to Build an Intelligent World”Your Trusted ICT Product Service Provider!
Sandy Yang/Global Strategy Director
WhatsApp / WeChat: +86 13426366826
Email: yangyd@qianxingdata.com
Website: www.qianxingdata.com/www.storagesserver.com
Business Focus:
ICT Product Distribution/System Integration & Services/Infrastructure Solutions
With 20+ years of IT distribution experience, we partner with leading global brands to deliver reliable products and professional services.
“Using Technology to Build an Intelligent World”Your Trusted ICT Product Service Provider!