---
title: "Deploy in the server room you already have"
description: "The platform deploys at single-server or small-cluster scale: one unit in an existing rack, or several as a compact GPU pool, without facility changes."
url: "https://www.lmtek.com/ai-hpc-servers/deployment"
locale: "en"
---

# Deploy in the server room you already have

The platform deploys at single-server or small-cluster scale: one unit in an existing rack, or several as a compact GPU pool, without facility changes.

| Specification | Value |
| --- | --- |
| Ambient, standard server room | 5–30°C |
| Coolant | 5–60°C |
| No facility plumbing | No CDU |
| Sustained chassis maximum | 8000 W |

A barebone platform: LM TEK supplies the chassis, the integrated cooling loop, and platform-level power. You specify the board, CPUs, GPUs, memory, and storage.

## A standard server room, stated precisely.

Eight full-height GPU slots in a four-rack-unit chassis is density that normally needs 6U or 8U. The closed in-chassis loop and in-chassis radiator mean it deploys without facility plumbing or an external CDU.

The claim is precise: no facility plumbing and no external CDU, in standard server-room conditions of 5 to 30°C ambient. In LM TEK torture testing at 30°C ambient, GPU package temperatures held at 76 to 81°C, inside the envelope a real server room can offer.

- Deploys in standard server-room conditions: 5 to 30°C ambient, 5 to 60°C coolant.
- No facility plumbing. No external CDU.
- Five 2000 W 80 PLUS Titanium PSUs with redundancy, up to 8000 W sustained.
- You still provision rack power and rack cooling headroom.

## Plan for the heat

A dense GPU server converts nearly all of its electrical draw into heat in the room. In LM TEK long-term endurance testing, a fully populated LM TEK 4U8GPU10KW measured 6800 to 7000 W sustained draw; that run has no written report yet. That heat load is defined by the GPUs and CPUs in the configuration, the same as any other max-spec 8-GPU server. Without added room cooling, ambient will rise under sustained load. The LM TEK 4U8GPU10KW’s operating envelope extends to 30°C ambient, so scope rack cooling for the configured load before deployment and keep the room inside that limit under sustained load.

Source: LM TEK long-term endurance testing.

Rear of the chassis under load, thermal camera: the heat the loop collects leaves into the room. Image from a separate test run (LM TEK test report 631), not the endurance test behind the draw figure.

## Four checks for the room.

- Verify rack power capacity before deployment. A fully-populated platform draws up to 8000 W sustained and requires adequate rack power provisioning.
- Confirm rack clearance against the dimensional drawing (760 × 482 × 178 mm) before rack integration.
- Operate within the stated envelope: 5–30°C ambient and 5–60°C coolant. Operation outside this range is not supported.
- Ask engineering to review any requirement outside the published envelope.

## Efficiency, and the water question

Power conversion uses 80 PLUS Titanium supplies, the highest rating in the certification; at multi-kilowatt sustained draw, conversion losses are a real operating cost and an increasingly common bid and tender criterion. And the water objection to AI infrastructure does not apply here: the loop is closed, filled once at assembly with engineered coolant, and consumes no water in operation. There is no facility water connection to consume from.

## Specifications and measured evidence.

[Full specifications](/ai-hpc-servers/4u8gpu10kw/specifications)

[Measured validation evidence](/ai-hpc-servers/4u8gpu10kw/validation)

### Full specifications

The datasheet, the operating limits and the dimensions for rack integration.

### Measured validation evidence

Torture-test and long-term thermals, each with its conditions and source.