ASRock Rack 1U4G-GENOA/2T 1U GPU Server

The ASRock Rack 1U4G-GENOA/2T is a compact 1U rackmount GPU server built on AMD EPYC Genoa (9004 series) processors, designed to deliver high-density AI and HPC compute performance within a standard single-rack-unit chassis. The platform supports up to four full-height, full-length GPU accelerators in a 1U form factor, a configuration made possible through a riser architecture that accommodates discrete PCIe Gen 5 cards. Dual 25GbE onboard networking via the /2T designation supports high-throughput data ingestion pipelines suited to inference serving and real-time analytics workloads.

The server is engineered around the AMD EPYC Genoa microarchitecture, which provides PCIe 5.0 lane density and DDR5 memory support, enabling low-latency communication between the CPU and attached GPU accelerators. This makes the platform well suited to AI inference at the edge and in space-constrained colocation environments where 1U density is a prerequisite. The combination of EPYC Genoa's large L3 cache, high memory bandwidth, and direct PCIe 5.0 GPU attachment reduces inference bottlenecks for transformer-based and convolutional neural network workloads.

Targeted at enterprise AI infrastructure teams, telecommunications edge deployments, and research institutions requiring GPU-accelerated compute without the footprint of a 2U or 4U chassis, the 1U4G-GENOA/2T occupies a distinct position in the ASRock Rack portfolio. Its 1U profile allows operators to maximize GPU density per rack unit in environments where power and space budgets are tightly managed, while the Genoa platform ensures compatibility with current-generation software stacks optimized for AMD's Zen 4 core architecture.

Ideal for

• AI inference serving at the network edge where rack space is limited to 1U per node
• High-density GPU cluster deployments in colocation facilities with strict power-per-rack constraints
• Real-time video analytics and computer vision pipelines requiring multiple GPU accelerators in compact form
• Telecommunications edge nodes running AI-assisted network optimization and traffic classification
• Scientific computing and HPC workloads leveraging EPYC Genoa's high core count and PCIe 5.0 bandwidth
• Enterprise private AI deployments where space-efficient GPU infrastructure underpins on-premises LLM inference

Technical specifications

Available from Omnixon Global. Submit an RFQ and our team will confirm configuration and availability for your order.