Microsoft Foundry Generative AI Chat App Practice Question 2: Data Zone Deployment Type
Microsoft Applied Skills: Develop a Generative AI Chat App Using the Microsoft Foundry SDK · Test 1 · question 2 of 10
A European bank is deploying a chat model in Microsoft Foundry. Compliance requires that prompts and responses be processed only inside the Azure EU Data Boundary, but the team does not want to be pinned to a single Azure geography and wants the higher default quota that comes with zone-wide routing. Traffic is bursty, and they refuse to commit to reserved capacity. Which deployment type should you choose?
Show answer and explanation
Answer:
Data Zone Standard (DataZoneStandard) routes traffic dynamically to datacenters within the Microsoft-defined data zone (US, EU, or APAC), and the EU data zone follows the Azure EU Data Boundary. It bills pay-per-token, so there is no capacity commitment, and the docs state it "provides higher default quotas than geography-based deployment types while keeping data within the specified zone." That satisfies every constraint in the stem.
- B is wrong because Global deployment types "may be processed in any Azure region." Choosing an EU region for the resource does not stop inference traffic from being routed outside the EU data boundary, which breaks the compliance requirement. - C is wrong because Standard processes within a single customer-specified Azure geography. It would satisfy residency, but the stem explicitly rules out being pinned to one geography and asks for the higher zone-wide quota. - D is wrong because provisioned types bill as reserved provisioned throughput units (PTUs). That is the reserved capacity commitment the team refuses, and it suits consistent high volume rather than the bursty traffic described.
Reference: Understanding deployment types in Microsoft Foundry Models