4U Server Case for NixOS Homelab

Dual GPU (Tesla P4 + RTX 3060) • AI/ML workload • Budget-first

✅ THE PICK

Chenbro RM41300-FS81

8 expansion slots, Tesla-GPU optimized, purpose-built for dual-GPU homelab compute. Best price for the job.

Chenbro RM41300-FS81

4U rackmount chassis. 7 full-length PCI-E slots (your 2 GPUs + room for future expansion). Steel/plastic construction, designed specifically for Tesla data-center cards but works equally well with consumer GPUs like your RTX 3060. Shipped with front 120mm + optional rear dual 80mm cooling fans.

Motherboard: EEB (12" × 13") — ATX compatible ✓

Expansion slots: 7 full-length

Depth: 21.5" (verify rack fit)

Height: 6.9" (standard 4U)

Cooling: 120mm front + 2× 80mm rear (opt.)

Power: Optional (bring your own)

$197.00 USD RackmountNet, in stock, fetched 2026-06-25

Where sold:
RackmountNet — $197.00 (in stock, Add to cart present)
Newegg — ~$244 (estimated, not live-verified)

Why it wins:

Cheapest verified price ($197) among all options
8 PCI-E slots = proven for dual-GPU AI/ML setups
Tesla-optimized = better documentation for your P4 + 3060 combo
Homelab community consensus: Chenbro quality > Rosewill at this price
Your PSU already fits (no add-on cost)

👍 Also Good

Rosewill RSV-R4000U

4U rackmount with 8 HDD bays + 7 PCI-E slots. More storage expansion than the Chenbro, but ~$50–100 more expensive. GPU clearance caps out at 320–400mm depending on configuration (both your cards fit, but less headroom). Slightly noisier stock fans. Consider if future NAS/storage expansion is a priority.

Estimated $300–350 USD (not live-verified)

SilverStone RM42-502

Premium 4U case. Liquid-cooling ready, better cable management, quieter. ~$500+. Overkill for a homelab hobby unless you plan serious sustained compute or water-cooling.

Estimated $500+ USD (not live-verified)

⚠️ Watch-outs

Depth (21.5"): Verify your rack/shelf is at least 24" deep. Will stick out slightly on shallower rigs.
No PSU included: You have one already ✓ so this doesn't cost you extra. Just confirm it fits ATX bracket inside the case.
Cooling for sustained AI workload: Stock fans are adequate for your cards (P4 = 50W, RTX 3060 ~170W peak). If running heavy CUDA training 24/7, consider adding an extra intake fan ($20–30).
Price verified live, 2026-06-25: Always confirm at checkout before committing.