How do you serve ML models across multiple regions?

hard

Answer

$(1) Deploy model to each region: latency to nearest user.$
$(2) Global load balancer routes by geography.$
$(3) Feature stores per region (or globally-replicated).$
$(4) Model registry syncs across regions.$
$(5) Consistency: eventually-consistent model versions OK; strong consistency for safety-critical.$
$(6) Failover: healthchecks + auto-route to backup region.$
$(7) Region-specific compliance (GDPR EU, data residency).$
$Multi-cloud (AWS + GCP) for resilience but complex.$

Check yourself — multiple choice

Random
Deploy per region (nearest user latency) + global LB + FS per region or replicated + registry sync + eventually-consistent versions + healthcheck failover + compliance (GDPR/residency); multi-cloud complex
One region
Not real

Multi-region: local deploy + LB + FS + registry sync + failover + compliance.

#infrastructure#deployment

215 interview questions in this topic.