Comparison
Availability zonevsCloud region
Availability zone
one datacentre lost power and a third of your instances went with it, which is the unit this is designed around.
An isolated datacentre or group of them within a region, with independent power, cooling and network, connected to its siblings over a low-latency link. It is the smallest unit of correlated failure a cloud offers, which makes it the unit you spread replicas across. Because the link between zones is fast but not free, spreading across them costs latency in single digits of milliseconds and real money in cross-zone data transfer.
Full entry →Cloud region
everything you run is in one geography, and a customer in another continent pays a hundred and fifty milliseconds for it.
A geographic grouping of availability zones, and the boundary most cloud services are scoped to. It determines user latency, which data residency laws apply, and what a regional outage costs you. Regions are far more independent than zones, which is what makes multi-region genuinely expensive: nothing is shared, so state has to be replicated over a slow, lossy, chargeable link.
Full entry →