Efficiency is migrating from ESG preference to procurement constraint. EU EED reporting → rating → minimum standards; Singapore DC-CFA2 allocation gate at PUE ≤ 1.25; US state mosaic (Virginia, Oregon, California). Vendor implications on UPS topology, PDU efficiency, cooling-electrical co-design.
Hall BESS as a distinct asset class. Wholesale grid-services asset deployed inside a data hall. Revenue stack across cost avoidance, ancillary services, capacity, demand response. Vendor set: Tesla Megapack, Fluence, Powin, Sungrow, Wärtsilä. Contract structure is the killer question.
Standalone deep-dive on the lithium UPS battery / BBU. LFP dominant, NMC niche, nickel-zinc pilot cohort. Killer thesis: pricing power migrates from UPS OEM to cell supplier over the decade. Trade-policy overlay (Section 301, 45X FEOC), 800V DC BBU migration, three-tier vendor set.
The Investment Layer scorecard flagged single-tenant concentration risk on the Oracle-OpenAI structure pre-close. In March 2026 that exact risk materialised. Oracle and OpenAI dropped the expansion; Microsoft stepped in to lease the dropped capacity. RESTRUCTURE, with realized-prediction validation.
The wired-recourse mechanic in market-tested form. High uncontracted principal at maturity, but CoreWeave's parent guarantee absorbs at Ba2/BB+ corporate credit. AMBER on the framework's 2x2, not RED. The distinction most credit coverage misses.
Walking the September 2024 Blackstone + CPP acquisition of AirTrunk through both AI infrastructure scorecards. LOW physical + LOW thesis + GREEN veto = UNDERWRITE (Premier). The calibration anchor for what a clean deal scores on the framework.