CASE 60 · PILLAR · 2026
An active-active payments API that runs everywhere, all the time.
A payments processor needed an active-active multi-region architecture for their core authorisation API — not warm-standby, not failover, but real concurrent serving from both regions with sub-second cross-region writes. We rebuilt the data layer on DynamoDB Global Tables and the routing on Route 53 latency records.
Payments processor
RELIABILITY
2026
RESULTS
What changed, by the numbers.
RTO (REGION LOSS)
< 30s
RPO
< 1s
CROSS-REGION CONFLICTS
0.0009%
TRAFFIC SPLIT
52 / 48
HOW IT WENT
The previous architecture was a "primary in us-east-1, warm-standby in us-west-2" setup. RTO had been 14 minutes in the most recent drill. The board had asked for "no observable downtime during a region event." That meant active-active.
DynamoDB Global Tables provided the multi-master data plane. Route 53 latency-based routing sent each request to the nearest healthy region, with health checks at the API Gateway endpoint level. EventBridge Global Endpoints replicated the event stream both ways. Idempotency keys ensured re-attempts didn’t double-charge.
Conflict rate landed at under 0.001% of writes, all auto-resolved by last-writer-wins on idempotent operations. During the quarterly drill we shut down us-east-1 traffic at the regional level; the system absorbed the redistribution in under 30 seconds with no customer impact. The board moved on to the next agenda item.
RELATED · SAME DOMAIN
Other engagements in this space.
READY WHEN YOU ARE
Let's get your AWS bill (and architecture) in order.
The discovery call is free. You walk away with at least one concrete idea — even if we never work together.