Oossa

Vertiv and NVIDIA unveil AI‑factory method promising up to 30% more token output

The new pre‑engineered, factory‑built approach could cut design time from 30 days to 3 and deployment time from 90 to 45 days.

NoteBy Published by Oossa: 1 min read

At NVIDIA’s GTC Berlin, Vertiv Holdings and NVIDIA announced an AI‑factory workflow that uses pre‑engineered reference designs and factory‑built hardware. The modeled scenario says the method can shrink configuration and virtual validation from about 30 days to three days, and halve on‑site deployment from 90 days to 45. With the same installed capacity, the faster rollout could generate up to 30% more cumulative AI token output in the first year. Polar Data Centers is applying the approach to the first 41 MW phase of its AI data center in Herøya, Norway.

Why it matters

Data‑center operators could bring AI compute online weeks earlier, potentially earning more revenue from AI workloads.

Was this article useful?
Share

Read next

Oossa · Newsletter

The week in AI, explained

Every Monday: the stories worth knowing, in plain language. Free, no spam.