About Us | Join Us
Background
COMPAL | AMD

Open, scalable CPU foundations plus Instinct GPU servers — air, DLC, and ORv3 options for AI and HPC.

Contact
Acceleration + compute

Acceleration + compute

Strong CPU foundation plus accelerator-ready GPU infrastructure — all within the AMD ecosystem

From inference to training

From inference to training

From general-purpose compute servers to high-density GPU accelerator platforms

Flexible thermal paths

Flexible thermal paths

Multiple cooling options matched to workload density and data centre constraints

SG720-2A

AMD EPYC™ 9005 Series

SG720-2A
  • AMD Instinct™ MI350X ready
  • Supports single & dual-phase DLC
  • PUE as low as 1.04

OG720-2A

AMD EPYC™ 9005 Series

OG720-2A
  • ORv3 - Open Rack Version
  • Memory Type: DDR5 RDIMM (up to 6400 MT/s)
  • 8 * AMD Instinct MI300X / MI325X OAM

SG223-2A-I

AMD EPYC™ 9004/9005 Series

SG223-2A-I
  • Supports up to 8 PCIe GPUs
  • Advanced immersion cooling
  • Optimized GPU thermal efficiency

SR425-1A

AMD EPYC™ 9005 Series

SR425-1A
  • Up to 200M IOPS performance
  • PCIe Gen6 E1.S storage
  • GPU Direct Storage ready

SX420-2A

AMD EPYC™ 9005 Series

SX420-2A
  • NVIDIA MGX™ architecture support
  • Flexible AI workload deployment
  • Large-scale memory expansion

SR224-2A

AMD EPYC™ 9005 Series

SR224-2A
  • Supports dual GPU configuration
  • Up to 500W CPU support
SG720-2A
MODEL SG720-2A
WORKLOAD AI training, inference, and GPU acceleration
FORM FACTOR 7U
PROCESSOR Dual AMD EPYC™ 9005 Series Processors
GPU AMD Instinct™ MI300X AMD Instinct™ MI325X AMD Instinct™ MI350X
MAX MEMORY 6TB
DRIVE BAY CAPACITY (MAX) 8 x NVMe U.2 + 2 x M.2
COOLING Air
MAX POWER(SYSTEM) 54V 3+3 Redundant, 4000W, Titanium PSU
RACKSTANDARD EIA 19”
Download
OG720-2A
MODEL OG720-2A
WORKLOAD AI training, inference, and GPU acceleration
FORM FACTOR 7OU
PROCESSOR Dual AMD EPYC™ 9005 Series Processors
GPU AMD Instinct™ MI300X AMD Instinct™ MI325X AMD Instinct™ MI350X
MAX MEMORY 6TB
DRIVE BAY CAPACITY (MAX) 8 x NVMe U.2 + 2 x M.2
COOLING Air
MAX POWER(SYSTEM) 54V 3+3 Redundant, 4000W, Titanium PSU
RACKSTANDARD OCP ORv3
Download
SG223-2A-I
MODEL SG223-2A-I
WORKLOAD AI training, inference, and GPU acceleration
FORM FACTOR 2U
PROCESSOR Dual AMD EPYC™ 9004/9005 Processor Family
GPU AMD Instinct™ MI210 NVIDIA H100 NVL AMD Instinct™ MI325X AMD Instinct™ MI350X
MAX MEMORY 6TB
DRIVE BAY CAPACITY (MAX) Up to 4 x E1.S + 2 x M.2
COOLING Immersion
MAX POWER(SYSTEM) 3200W 1+1 Redundant, Titanium CRPS PSU
RACKSTANDARD EIA 19”
Download
SR425-1A
MODEL SR425-1A
WORKLOAD High-capacity and data-intensive storage
FORM FACTOR 4U
PROCESSOR Single AMD EPYC™ 9005 Series Processors
GPU NVIDIA H200 NVL NVIDIA RTX PRO™ 6000 Blackwell Server Edition NVIDIA RTX PRO™ 4500 Blackwell Server Edition
MAX MEMORY 4TB
DRIVE BAY CAPACITY (MAX) 32 x E1.S
COOLING Air
MAX POWER(SYSTEM) 3200W N+1 Redundant, Titanium PSU
RACKSTANDARD EIA 19”
Download
SX420-2A
MODEL SX420-2A
WORKLOAD AI training, inference, and GPU acceleration
FORM FACTOR 4U
PROCESSOR Dual AMD EPYC™ 9005 Series Processors
GPU NVIDIA H200 NVL NVIDIA L40S GPU NVIDIA RTX PRO™ 6000 Blackwell Server Edition NVIDIA RTX PRO™ 4500 Blackwell Server Edition
MAX MEMORY 8TB
DRIVE BAY CAPACITY (MAX) 16 x E1.S or 12 x U.2 + 2 x M.2
COOLING Air
MAX POWER(SYSTEM) 3200W 3+1 Redundant, Titanium CRPS PSU
RACKSTANDARD EIA 19”
Download
SR224-2A
MODEL SR224-2A
WORKLOAD Cloud, virtualization, and enterprise applications
FORM FACTOR 2U
PROCESSOR Dual AMD EPYC™ 9005 Series Processor Family
GPU NVIDIA H100 GPU NVIDIA L40S GPU
MAX MEMORY 6TB
DRIVE BAY CAPACITY (MAX) Max: front 24 x 2.5" or 12 x 3.5" + rear 2 x 2.5" + 2 x M.2
COOLING Air
MAX POWER(SYSTEM) 2400W 1+1 Redundant, Titanium PSU
RACKSTANDARD EIA 19”
Download
OX420-2A
MODEL OX420-2A
WORKLOAD AI training, inference, and GPU acceleration
FORM FACTOR 4OU
PROCESSOR Dual AMD EPYC™ 9005 Series Processors
GPU NVIDIA H200 NVL NVIDIA L40S GPU NVIDIA RTX PRO™ 6000 Blackwell Server Edition NVIDIA RTX PRO™ 4500 Blackwell Server Edition
MAX MEMORY 32TB
DRIVE BAY CAPACITY (MAX) 8 x E1.S + 2 x M.2
COOLING Air
MAX POWER(SYSTEM) ORV3 48V IT Gear cable
RACKSTANDARD OCP ORv3
Download
SG520-2A-L1
MODEL SG520-2A-L1
WORKLOAD AI training, inference, and GPU acceleration
FORM FACTOR 5U
PROCESSOR Dual AMD EPYC™ 9004 Series Processor family
GPU AMD Instinct™ MI355X
MAX MEMORY 6TB
DRIVE BAY CAPACITY (MAX) 8 x U.2 2.5" + 2 x M.2
COOLING DLC (1-Phase)
MAX POWER(SYSTEM) 4000W 54V 3+3 Redundant, Titanium CRPS PSU
RACKSTANDARD EIA 19”
Download
SD223-8A-L1
MODEL SD223-8A-L1
WORKLOAD HPC and scale-out CPU compute
FORM FACTOR 2U
PROCESSOR Dual AMD EPYC™ 9004 Series Processor family
GPU N/A
MAX MEMORY 24TB
DRIVE BAY CAPACITY (MAX) Per node: up to 6 x 2.5" + 2 x M.2; chassis: up to 24 x 2.5" + 8 x M.2
COOLING DLC (1-Phase)
MAX POWER(SYSTEM) 2000W 2+1, 2+2, 3+1 Redundant, Platinum/Titanium PSU
RACKSTANDARD EIA 19”
Download
SR124-2A
MODEL SR124-2A
WORKLOAD Cloud, virtualization, and enterprise applications
FORM FACTOR 1U
PROCESSOR Dual AMD EPYC™ 9005 Series Processor Family
GPU N/A
MAX MEMORY 6TB
DRIVE BAY CAPACITY (MAX) Max: front 12 x 2.5" + 4 x 3.5" + 2 x M.2
COOLING Air
MAX POWER(SYSTEM) 1600W 1+1 Redundant, Platinum/Titanium PSU
RACKSTANDARD EIA 19”
Download
SD223-8A
MODEL SD223-8A
WORKLOAD HPC and scale-out CPU compute
FORM FACTOR 2U
PROCESSOR Dual AMD EPYC™ 9004 Series Processor family
GPU N/A
MAX MEMORY 24TB
DRIVE BAY CAPACITY (MAX) Per node: up to 6 x 2.5" + 2 x M.2; chassis: up to 24 x 2.5" + 8 x M.2
COOLING Air
MAX POWER(SYSTEM) 2000W 3+1 Redundant, Platinum/Titanium PSU
RACKSTANDARD EIA 19”
Download
SR223-1A
MODEL SR223-1A
WORKLOAD Cloud, virtualization, and enterprise applications
FORM FACTOR 2U
PROCESSOR Single AMD EPYC™ 9005 Series Processors Family
GPU N/A
MAX MEMORY 3TB
DRIVE BAY CAPACITY (MAX) Max: front 24 x 2.5" or 12 x 3.5" + rear 2 x 2.5" + 2 x M.2
COOLING Air
MAX POWER(SYSTEM) 2400W 1+1 Redundant, Platinum/Titanium PSU
RACKSTANDARD EIA 19”
Download
SD223-8A-L2
MODEL SD223-8A-L2
WORKLOAD HPC and scale-out CPU compute
FORM FACTOR 2U
PROCESSOR Dual AMD EPYC™ 9004 Series Processor family
GPU N/A
MAX MEMORY 24TB
DRIVE BAY CAPACITY (MAX) Per node: up to 6 x 2.5" + 2 x M.2; chassis: up to 24 x 2.5" + 8 x M.2
COOLING DLC (2-Phase)
MAX POWER(SYSTEM) 2000W 2+1, 2+2, 3+1 Redundant, Platinum/Titanium PSU
RACKSTANDARD EIA 19”
Download
SD225-2A
MODEL SD225-2A
WORKLOAD HPC and scale-out CPU compute
FORM FACTOR 2U
PROCESSOR Dual AMD EPYC™ 9004/9005 Series Processor Family
GPU NVIDIA H200 NVL
MAX MEMORY 12TB
DRIVE BAY CAPACITY (MAX) Per node: up to 12 x E1.S; chassis: up to 24 x E1.S
COOLING Air
MAX POWER(SYSTEM) 3200W 1+1 Redundant, Platinum/Titanium PSU
RACKSTANDARD EIA 19”
Download
SR221-2A
MODEL SR221-2A
WORKLOAD Cloud, virtualization, and enterprise applications
FORM FACTOR 2U
PROCESSOR Dual AMD EPYC™ 9004/9005 Series Processor Family
GPU NVIDIA H100 GPU NVIDIA L40S GPU
MAX MEMORY 6TB
DRIVE BAY CAPACITY (MAX) Max: front 24 x 2.5" or 12 x 3.5" + rear 2 x 2.5" + 2 x M.2
COOLING Air
MAX POWER(SYSTEM) 2400W 1+1 Redundant, Titanium PSU
RACKSTANDARD EIA 19”
Download
SR121-2A
MODEL SR121-2A
WORKLOAD Cloud, virtualization, and enterprise applications
FORM FACTOR 1U
PROCESSOR Dual AMD EPYC™ 9004/9005 Series Processor Family
GPU N/A
MAX MEMORY 6TB
DRIVE BAY CAPACITY (MAX) Max: front 12 x 2.5" + 4 x 3.5" + 2 x M.2
COOLING Air
MAX POWER(SYSTEM) 1600W 1+1 Redundant, Platinum/Titanium PSU
RACKSTANDARD EIA 19”
Download
SR112-2A
MODEL SR112-2A
WORKLOAD Cloud, virtualization, and enterprise applications
FORM FACTOR 1U
PROCESSOR Dual AMD EPYC™ 7003 Milan Series Processors Family
GPU N/A
MAX MEMORY 4TB
DRIVE BAY CAPACITY (MAX) 10 x 2.5"
COOLING Air
MAX POWER(SYSTEM) 800W 1+1 Redundant, Platinum/Titanium PSU
RACKSTANDARD EIA 19”
Download
SR212-2A
MODEL SR212-2A
WORKLOAD Cloud, virtualization, and enterprise applications
FORM FACTOR 2U
PROCESSOR Dual AMD EPYC™ 7003 Milan Series Processors Family
GPU N/A
MAX MEMORY 4TB
DRIVE BAY CAPACITY (MAX) 24 x 2.5"
COOLING Air
MAX POWER(SYSTEM) 1600W 1+1 Redundant, Platinum/Titanium PSU
RACKSTANDARD EIA 19”
Download

Service

Workload fit (training vs inference vs HPC)
Cooling integration readiness (air/DLC)
ORv3 integration support

Get the latest on AI infrastructure

Contact