Early sight of capacity
Marketing Services Agreements with operators across our markets mean we often see new allocation and replacement space early, with first preference.
AI buyers increasingly ask for megawatts, at densities much of the existing estate was never built to carry, on dates that do not move. We broker the capacity that can actually take it.
When we ask an operator whether a 10U air-cooled 8× B200 node at about 14 kW will work in their hall, we already know what the answer needs to be, because we ship the node.
Marketing Services Agreements with operators across our markets mean we often see new allocation and replacement space early, with first preference.
Air, rear-door heat exchanger, direct-to-chip liquid, in-row CDU. We qualify halls on what they can carry, not what the brochure says.
We do not shortlist Tier II facilities for production AI workloads.
Availability, pricing and terms are facility-specific and move constantly. Nothing here is an offer or a guarantee of capacity.
Under a Marketing Services Agreement, the operator pays our fee, and we tell you before shortlisting. Where you engage us directly for advisory or a supply-and-host delivery, the fee is agreed in writing before we start.
Mostly 40 kW to 165 kW+ per rack, air and direct-to-chip liquid. We also handle conventional enterprise colocation.
In our representative 256-node B200 example, about 3.6 MW is IT load and about 5.2 MW is total facility power once cooling and electrical overhead are added. We force that distinction into writing in every RFI.
The United States, Malaysia and Indonesia, with wider ASEAN coverage through partner facilities.