Methodology.

Primary evidence

We prioritize official technical documentation, institutional resources, dated benchmark publishers and company statements. Company announcements are evidence of what a vendor says, not independent product validation. Source dates and material limits accompany the research.

Planning models

The Scaling Lab is an Amdahl-style fixed-work model with an explicit overhead assumption. The Facility Planner uses a constant average electrical-load factor and PUE over a selected period. Both omit variables a real engineering study would need.

Evidence completeness

The Project Brief counts non-empty evidence fields and separately lists critical missing fields. It does not verify the truth, sufficiency, approvals or technical validity of entered content.

Independent analysis

Workload patterns and future platform uses are original educational interpretations. No company listing implies commercial involvement. No live prices, capacity availability or buyer interest is asserted.