Kakao, Data Center Fire, the Data Residency Dilemma

Erm, i don't know...if I were architecting the backup locations (yes, plural) for that data center in question, even if I can't reach some ideal long disatance (like the 1,100 miles mentioned), i would still opt for standing up a backup location - even if it was only a few hundred miles away...something not-so-ideal is still better than nothing. Or, maybe there were backup locations, but you know, someone failed to go through their restore/disaster recovery at other location procedures and processes? If i were the leadership, i'd be asking all sorts of questions beginning with: who cares if the backup is not 1,100 miles away, why didn't you at least have some/any back up location1?!

Microsoft Azure doesn't agree with this author. They have 3 Availability Zones in their Korea Central region.

It seems like there are a large number of commercial data centres in operation in South Korea, and not all in Seoul: https://www.datacenterdynamics.com/en/news/?tag=south-korea

Kakao's incident was caused by a single building fire. It's quite a leap to suggest it is impossible to deliver a resilient service in a country the size of South Korea.

Data center locations are chosen primarily for latency and cost (mostly power) reasons, not for disaster recovery reasons. A second datacenter near Busan would work fine.

Disaster recovery planning for "what if something takes out Seoul" obviously requires backups *outside* of the country.

Kinda funny the gov is calling Kakao a monopoly, yet not the chaebols?