Releases in this digest

Headlines link to our pages; synopses are MediaRelease.co summaries.

STTGDC to host Neysa’s AI Compute Infrastructure at Noida Data Centre

STTGDC · Original source

India-based AI cloud company Neysa has selected STTGDC’s STT Noida 2 data centre to host GPU infrastructure and its Velocis AI cloud platform. The first phase of the multi-year relationship will support up to 9,000 GPUs for high-density AI training and inference and is expected to go live from the first quarter of calendar 2027. The deployment is aimed at production AI workloads requiring data residency in India.

Crusoe Announces Support for NVIDIA DSX POD, Bringing Field-Proven Crusoe Spark Platform to the Expanding Modular Ecosystem

Crusoe · Original source

Crusoe says its Crusoe Spark modular AI data center platform will support NVIDIA’s new DSX POD reference design for modular NVIDIA Vera Rubin deployments. Crusoe has built and operated modular data centers since 2020 and plans mass production at its 352,000-square-foot Spark factory in Brighton, Colorado. Head of Spark Bryan Schramm said the company can deliver productive AI capacity in as little as three months.

Vertiv targets up to 30% more cumulative AI token output in first 12 months using NVIDIA DSX, Polar deploys approach at Norway site

Vertiv · Original source

Vertiv and NVIDIA introduced an AI factory approach at NVIDIA GTC Berlin that pairs pre-engineered, virtually validated reference designs with factory-built power and cooling infrastructure. In Vertiv’s modeled scenario, design and deployment stages fall from about 120 days to 48, bringing capacity online roughly 10 weeks sooner and yielding up to 30% more token output in the first year. Polar Data Centers is applying the approach to the first 41 MW phase of its Herøya, Norway campus.

Crusoe Introduces Crusoe Spark Liquid, Its Next-Generation Modular AI Factory to Power Crusoe Cloud

Crusoe · Original source

Crusoe has introduced Crusoe Spark Liquid, the sixth generation of its prefabricated modular AI data center. Each unit supports up to 3.5 MW of compute using closed-loop, non-evaporative direct liquid cooling, a fivefold density increase over earlier generations, and is built for NVIDIA GB300 NVL72 and Vera Rubin NVL72 systems. Production will join existing Spark manufacturing at Crusoe’s Brighton, Colorado factory in 2027. Ollama is cited as a current Spark customer.

Equinix Opens Innovation Centers Globally to Accelerate Enterprise Infrastructure Deployments

Equinix · Original source

Equinix has launched Enhanced Performance Innovation Centers (EPIC), a network of production-ready liquid cooling environments inside live Equinix data centers. Enterprises can trial liquid-cooled GPU and CPU platforms with hardware partners and test AI training and inference workloads before scaling. EPIC is live in Frankfurt, London, Osaka, Paris and Sydney, with more locations expected in 2027. Criteo is cited as a customer that tested liquid cooling with Equinix before scaling.

Vantage and Heron Power Electronics Announce Gigawatt-Scale 800 VDC Agreement to Accelerate AI Infrastructure Delivery

Vantage · Original source

Vantage has secured more than 1 GW of manufacturing capacity for Heron Power Electronics’ Heron Link solid-state transformer to deploy across its hyperscale data center campuses. Heron Link converts medium-voltage grid power directly to 800-volt direct current with integrated energy storage, targeting more than 98.5% efficiency. The companies say the design cuts conversion stages and equipment compared with 480 VAC systems and supports megawatt-class AI rack densities.

Want these digests by email?

We are gauging interest in a daily or weekly newsletter for each area. Tell us which areas you would read.

Register interest