What’s supposed to happen to a data centre’s power and cooling systems during a significant earthquake? For a facility built to standard commercial specifications, the honest answer is that nobody’s entirely sure, because that scenario was never part of the original design brief. For a facility built to mission-critical standards, the answer is supposed to be nothing: the systems keep running. That difference, between a building that’s expected to survive an event and one that’s expected to keep operating through it, is most of what the term actually describes.
At a Glance: Two Design Standards, Two Different Assumptions
| Consideration | Standard commercial design | Mission-critical design |
| Structural approach | Meets local building code minimums | Engineered for continued operation during the event itself |
| Backup power | Covers planned outages and brief grid interruptions | Sized and tested for extended, unplanned grid loss |
| Maintenance access | Some systems require scheduled downtime to service | Concurrently maintainable, serviced without taking the facility offline |
Indonesia’s Seismic Risk Isn’t a Footnote in the Design Brief
A data centre in Indonesia has to account for the country’s position along the Pacific Ring of Fire from the outset, not as an afterthought bolted onto a template built for a calmer market. K2’s own first large-scale campus in the country, developed through a joint venture with Sinar Mas Land in Greater Jakarta, was designed around four planned facilities delivering close to 60 megawatts of IT capacity, with structural and power resilience built into the base design rather than added after the fact. A second site, on land acquired in Karawang, extends that same approach to a separate location rather than concentrating all of the country’s capacity in one place.
Spreading capacity across more than one site isn’t just a growth strategy; it’s also a resilience decision. A single campus, however well engineered, still represents a single location, and a tenant relying entirely on one site in a seismically active country is accepting a concentration of risk that a second, independently sited facility can meaningfully reduce. Where a workload can be split or mirrored across both locations, a localised event affecting one site doesn’t have to mean a full outage for the tenant relying on it.
Redundancy Isn’t Optional Once a Facility Is Actually Called Mission Critical
The term mission critical facilities gets used loosely across the industry, but it has a fairly specific engineering meaning: redundant power paths, redundant cooling, and a design that allows components to be serviced without interrupting the systems a tenant depends on. In a market where grid stability can’t always be assumed, that redundancy isn’t a premium feature; it’s closer to a baseline requirement for any tenant running workloads that genuinely can’t afford downtime.
- N+1 redundancy: at least one spare unit beyond what’s needed to run at full load
- 2N redundancy: a fully duplicated, independent backup system running in parallel
- Concurrent maintainability: the ability to service or replace equipment without a shutdown
Does More Redundancy Just Mean More Equipment Sitting Idle?
Not quite. The additional systems aren’t simply spares gathering dust; they’re part of how the facility is tested and operated day to day, since a redundant path that’s never exercised isn’t one a tenant can actually trust in an emergency. Regular failover testing, where load is deliberately shifted onto backup systems under controlled conditions, is as much a part of mission-critical operations as the hardware itself, and it’s one of the more meaningful differences between a facility that claims the designation and one that’s actually built and run to it.
None of this is unique to Indonesia specifically, since the same engineering standards apply anywhere a tenant’s workload genuinely can’t tolerate an outage. What Indonesia does is make the gap between a standard build and a mission-critical one more visible, since the conditions that redundancy is designed for show up there more often than in markets with calmer grids and quieter fault lines. A tenant evaluating a site in this kind of market is really asking a narrower question than it might seem: not whether the building will still be standing afterwards, but whether the systems inside it were ever expected to keep running while it happened.
K2 Strategic designs its Indonesia campuses to the same mission-critical standard it applies everywhere else. Contact K2 Strategic to talk through what that standard actually covers.