Zhivko Todorov
ALL CASE STUDIES

CASE 60 · PILLAR · 2026

ACTIVE-ACTIVEMULTI-REGIONDYNAMODB GLOBALROUTE 53

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.

INDUSTRY

Payments processor

DOMAIN

RELIABILITY

DELIVERED

2026

STACK

DYNAMODB GLOBAL TABLES·ROUTE 53 LATENCY ROUTING·API GATEWAY·LAMBDA·EVENTBRIDGE GLOBAL ENDPOINTS·CLOUDWATCH SYNTHETICS

RESULTS

What changed, by the numbers.

RTO (REGION LOSS)

< 30s

AUTOMATIC HEALTH CHECKS

RPO

< 1s

GLOBAL TABLE REPLICATION

CROSS-REGION CONFLICTS

0.0009%

OF WRITES, AUTO-RESOLVED

TRAFFIC SPLIT

52 / 48

NATURAL BY LATENCY

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.

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.

Or email directly →