4 GPU PCIe AI server, 4U
Four PCIe GPUs in a 4U chassis with generous airflow, 24 drive bays and room for expansion cards, for teams that want storage and GPUs in one node.
GPU: NVIDIA RTX PRO 6000 Blackwell Server Edition, H200 NVL or L40S
- GPUs
- 4 x RTX PRO 6000 Blackwell Server Edition, H200 NVL or L40S, with slots for 2 more later
- CPU
- 2 x AMD EPYC 9005 or 2 x Intel Xeon 6
- Memory
- Up to 2 TB DDR5 RDIMM
- Storage
- 2 x M.2 boot, up to 24 x 2.5 inch NVMe or SATA bays
- Networking
- 2 x 400 Gb/s ConnectX-7, 2 x 25 GbE, spare PCIe slots for more adapters
- Power
- Redundant PSUs, about 3.5 kW per node planning figure
Overview
The 4U four GPU server trades density for flexibility. The taller chassis has quieter fans, more PCIe slots and up to 24 hot swap drive bays, so one node can hold the GPUs, a local dataset of several hundred terabytes and extra adapters. Two more GPU slots are available in some layouts for growth.
It is a strong fit for departmental AI systems, lab rooms, edge sites and anywhere a rack full of 2U nodes is not the goal. Teams building a data lake and inference service on one box, or running simulation and rendering alongside AI, get the GPU options of the 2U class with much more room around them. A tower conversion kit lets it stand under a desk.
Processor, memory and GPU options match the 2U class: dual AMD EPYC 9005 or Intel Xeon 6, DDR5 to 2 TB, RTX PRO 6000 Blackwell Server Edition, H200 NVL or L40S. Power supplies can be specified for 120 V sites with L40S cards, or 200 to 240 V for full 600 W Blackwell loads.
Use cases
- Departmental AI servers
- Storage and GPU in one node
- Edge and lab deployments
- Simulation and rendering
- Video analytics
Specifications
| GPUs | 4 x NVIDIA RTX PRO 6000 Blackwell Server Edition, H200 NVL or L40S, expandable to 6 in some layouts |
|---|---|
| GPU interconnect | PCIe Gen5 x16 per GPU; NVLink bridges with H200 NVL |
| CPUs | 2 x AMD EPYC 9005 series or 2 x Intel Xeon 6 |
| System memory | 24 or 32 x DDR5 RDIMM, up to 2 TB |
| Boot | 2 x M.2 NVMe, mirrored |
|---|---|
| Data | Up to 24 x 2.5 inch bays, NVMe or SATA, hot swap |
| RAID | Hardware RAID or HBA for the drive bays |
| GPU fabric | 2 x 400 Gb/s ConnectX-7 or 2 x 200 Gb/s |
|---|---|
| Frontend | 2 x 25 GbE onboard, optional 100 GbE |
| Management | Dedicated 1 GbE BMC port, TPM 2.0 |
| Form factor | 4U rackmount, air cooled, tower conversion kit available |
|---|---|
| Power supplies | Redundant 2,000 W class PSUs, 200 to 240 V input; 1,600 W class for 120 V sites |
| Planning power | About 3.5 kW per node |
| Acoustics | Lower fan speed than the 2U class, suitable for lab rooms |
Figures describe the system class. Exact configuration, weights and power draw are confirmed on the quotation.
What's included
- Rails, power cords and cable management arm
- Burn in and acceptance test report
- Operating system imaging and firmware baseline on request
- 3 year warranty, options to 5 years with advance replacement
- NVIDIA end user registration handled with the order
Compliance
Related
Related systems
4 GPU PCIe AI server, 2U
- GPUs
- 4 x RTX PRO 6000 Blackwell Server Edition
- CPU
- 2 x AMD EPYC 9005 or 2 x Intel Xeon 6
- Memory
- Up to 1.5 TB DDR5 RDIMM
- Chassis
- 2U, about 3.5 kW, air cooled
Rackmount 4U workstation, up to 4 GPUs
- GPUs
- Up to 4 x RTX PRO 6000 Blackwell Max-Q
- CPU
- AMD Ryzen Threadripper PRO or 1 x Intel Xeon 6
- Memory
- 8 channel DDR5 RDIMM, 128 GB to 2 TB
- Chassis
- 4U, about 1.8 kW, air cooled
2U 24 bay all NVMe storage server
- CPU
- 2 x AMD EPYC 9005 or 2 x Intel Xeon 6
- Memory
- 24 x DDR5 RDIMM, 512 GB to 1.5 TB
- Software
- Parallel file systems, NVMe over Fabrics
- Chassis
- 2U, about 1.3 kW, air cooled
4 GPU RTX 5090 AI server, 4U
- GPUs
- 4 x NVIDIA GeForce RTX 5090, 32 GB GDDR7 each
- CPU
- 1 or 2 x AMD EPYC 9005 or Intel Xeon 6
- Memory
- Up to 1.5 TB DDR5 RDIMM
- Chassis
- 4U, about 3.5 kW, air cooled
8 GPU RTX 5090 AI server, 4U
- GPUs
- 8 x NVIDIA GeForce RTX 5090, 32 GB GDDR7 each
- CPU
- 2 x AMD EPYC 9005 or 2 x Intel Xeon 6
- Memory
- Up to 2 TB DDR5 RDIMM
- Chassis
- 4U, about 6.5 kW, air cooled
Related guides
Buying data center GPUs, export and registration
The compliance steps that come with every HGX, H200 NVL, Server Edition and NVL72 purchase, explained so your order ships on time.
In this guide
- Why a GPU purchase comes with paperwork
- Which products are covered
Choosing a GPU server for LLM inference
A practical method for sizing an inference server from the model, the quantization and the concurrency you expect, with worked examples.
In this guide
- Start with the model, not the server
- Step 1: how much GPU memory the weights need
HGX vs PCIe GPU servers
The real differences between an HGX node and a PCIe GPU server, and a simple rule for choosing between them.
In this guide
- Two kinds of 8 GPU server
- The interconnect is the point
How to request a quotation
A short guide to getting a complete, accurate quotation, from the details we need to what the quotation contains.
In this guide
- Why there are no prices on this site
- Three ways to ask
Compare it with the next class up or down
The nearest classes we quote are the 4 GPU PCIe AI server, 2U and the Rackmount 4U workstation, up to 4 GPUs. If neither is quite right, tell us the workload and we specify the system between them.
- (858) 613-1818
- sales@computermodules.com
- Monday to Friday, 8am to 5pm Pacific