Reference

Β·

Supporting source

Β·Preview

AI/HPC per-rack power density

30-50+ kW

As ofFY2024Β·Sources3Β·Supporting

AI/HPC workload per-rack power density runs 30-50+ kW versus the 6-10 kW traditional enterprise baseline β€” breaking air-cooled CRAC/CRAH economics above ~25 kW and forcing liquid cooling adoption.

Compare plans β†’

Figure in context

AI/HPC workload per-rack power density runs 30-50+ kW versus the 6-10 kW traditional enterprise baseline β€” breaking air-cooled CRAC/CRAH economics above ~25 kW and forcing liquid cooling adoption.

AI/HPC workload per-rack power density runs 30-50+ kW versus the 6-10 kW traditional enterprise baseline β€” breaking air-cooled CRAC/CRAH economics above ~25 kW and forcing liquid cooling adoption.

Time scope

FY2024

Source basis

Supporting source

Interpretation notes

What this tells you

AI/HPC workload per-rack power density runs 30-50+ kW versus the 6-10 kW traditional enterprise baseline β€” breaking air-cooled CRAC/CRAH economics above ~25 kW and forcing liquid cooling adoption.

What not to do with it

Use the linked report for interpretation and keep basis differences explicit.

Related figures

Adjacent numbers that add context without drowning the value.

Report context

Atlas actors in this figure's reports

Profiles covered in the report that cite this number.

AI/HPC per-rack power density Β· Insight