Resiliency Considerations with Direct Liquid Cooling

Overview

This document is the report Uptime Intelligence published in November 2023 as UII Briefing Report 117. The full report is 14 pages and needs registration or a subscription; the public part is a 4-page executive summary. What is described here rests only on the landing-page description and abstract. The report's detailed findings, diagrams and recommendations are outside the scope of this page. This page is therefore a preview rather than a substitute for the full text; for detailed assessment, consult the report itself on the publisher's site. This page is therefore a preview rather than a substitute for the full text; for detailed assessment, consult the report itself on the publisher's site.

The starting point is that direct liquid cooling is expanding beyond technical computing into mission-critical installations that have strict cooling resiliency demands. The report covers both cold-plate and immersion systems. According to the abstract, DLC introduces new failure scenarios that conventional air cooling does not have, and it complicates the split of responsibilities between facilities and IT teams. The abstract treats this as a design question, not only an operational one.

The authors note that applicable guidance is scarce. To help fill the gap, the report proposes guiding questions that start from business objectives. In other words, instead of a ready-made specification it offers a framework in which each site questions its resiliency level against its own continuity target. The document is not a standard and contains no numerical design criteria.

Key points

1. UII Briefing Report 117, November 2023; the 4-page executive summary of the 14-page report is public.
2. Covers cold-plate and immersion DLC.
3. DLC is expanding beyond technical computing into mission-critical sites.
4. DLC introduces new failure scenarios.
5. It complicates the split of responsibilities between facilities and IT teams.
6. Applicable guidance is scarce; the report proposes guiding questions starting from business objectives.

Using it in design

• DLC design checks: CDU redundancy, TCS loop isolation and ride-through for pumps.
• List the failure scenarios in the cooling loop (leak, pump loss, CDU failure) at design stage and assess the impact of each.
• Define the responsibility boundary of the liquid loop between facilities and IT explicitly in contracts and operating procedures.
• Derive the resiliency level from the business objective (acceptable downtime) first, then choose equipment redundancy to match.

Who should read it

Cooling designers and operations teams planning to add liquid cooling to a mission-critical site.

Publication details

Authors

Uptime Institute Intelligence

Publisher

Uptime Institute Intelligence

Year

2023

Document number

UII Briefing Report 117

Document language

English

Source type

Industry body

Access

Citation only

Licence

Copyright retained; citation only — the publisher’s terms apply to the original document

Citation