STT GDC Noida 1 is 17 MW on the facility card and 35 MW in a separate narrative on the same operator site. Both figures are published, neither has been withdrawn, and the 18 MW between them is larger than every Mumbai city building the same company runs. The building is under construction. No rack count, floor area, PUE or delivery date is published. This registry keeps 17 MW — the card value, because cards are what reconcile — and records the 35 as a scalability claim, because on a building that does not yet exist, scalability is a design intention rather than a capacity.
1At a glance
2Which number governs
Three readings of a 17 and a 35 on the same page, and no way to choose between them from outside the company.
| If it means | Then | Consequence |
|---|---|---|
| Phase one is 17 MW, the shell can reach 35 | both are true, on a schedule nobody has published | Noida 1 is a 35 MW building with a 17 MW opening phase |
| 17 MW is what is contracted, 35 is design | the card is the commercial figure | the portfolio headline should use 17 |
| 17 is an error and 35 is correct | the Noida portfolio is 85 MW, not 63 | the headline is 22 MW wrong |
The first reading is the most likely, and it is also the most useful: a phased building where the first phase is the sellable capacity and the design ceiling is twice that. That is standard practice and entirely reasonable.
The problem is that a scalability figure and a capacity figure look identical in a table. If 35 MW enters a market model, Noida 1 becomes the largest building in the portfolio and the Noida headline becomes wrong. Nothing on the page stops that happening.
3Why the lower one is the right one to hold
The choice of 17 over 35 is a methodology decision, and it should be stated rather than assumed.
| Reason | Detail |
|---|---|
| Cards reconcile | Across eight STT cities the cards sum to 389.5 MW; the headlines to 354. Cards are the checkable layer |
| Scalability is not capacity | A ceiling describes what a site could hold, not what will be energised |
| The building is under construction | Nothing has been commissioned, so neither figure describes existing supply |
| Conservative arithmetic | Understating a portfolio is recoverable. Overstating it contaminates every ratio built on it |
| The gap is recorded | the 18 MW is published here, not deleted |
The same rule is applied consistently across this file. CtrlS’s 700 MW Chandanvelly scalability is not counted, STT’s 400 MW Palava design capacity is not counted, and the 1.2 GW and 1.4 GW ceilings on the two CtrlS Hyderabad parks are not counted. Each one is published on its own page as a claim.
That consistency is the point: a buyer can disagree with the choice, but they cannot claim the numbers were inflated.
4Procurement
5Limits and open questions
- Two capacities, no reconciliation. The core defect of this record.
- No dates. Under construction with no start or completion.
- No racks, floor area or PUE. Density cannot be computed.
- No basis stated for either the 17 or the 35, though the city headline is labelled IT capacity.
- Single source. No trade press or directory carries either figure.
- What would close it. One sentence from the operator saying which figure describes what.
6Sources
Sources: STT GDC India Noida city page, STT GDC India facility pages for Noida, STT GDC India corporate site, and Indian market commentary on Noida data centre supply.