PRODUCTION SCENARIO
A bank in Germany runs an assistant on Amazon Bedrock that is throttled during peak hours in a single Region. The bank's regulator requires that inference data be processed only within Europe.
Which solution will meet these requirements?
Answering here is anonymous. Nothing is saved unless you sign in.
Show answer and explanation
Answer: Use a geographic cross-Region inference profile for Europe.
Cross-Region inference spreads requests across Regions to absorb peaks, and AWS says to choose a geographic profile when you have data residency requirements, because it routes only to Regions inside the chosen geography. A global profile can route to any commercial Region worldwide, breaking the regulator's condition. Provisioned Throughput cannot be attached to an inference profile at all. A second stack in US East adds capacity but processes data outside Europe.