You save $69.98
AZ-305 Premium Bundle
- Premium File 382 Questions & Answers
- Last Update: Nov 20, 2026
- Training Course 87 Lectures
- Study Guide 933 Pages
You save $69.98
Architecture review: Publish subscribe topics and Event Grid notification. Requirement A: Each order event must be processed by billing, shipping and analytics independently. Requirement B: A storage blob upload should notify multiple Azure Functions through filterable event subscriptions. Which complete two-part proposal satisfies both constraints?
For requirement A, topics fan out a message to independently managed subscriptions with durable handling. For requirement B, event Grid distributes discrete source events to configured subscribers. Verify both independently: All three subscribers receive an order event without direct calls from checkout; Only event handlers matching the subject filter are invoked for the test upload.
The option covers one requirement but substitutes Web PubSub real time for Publish subscribe topics. The missing acceptance result is: All three subscribers receive an order event without direct calls from checkout.
Topics fan out a message to independently managed subscriptions with durable handling. Event Grid distributes discrete source events to configured subscribers. The proposed steps (use Azure Web PubSub with authenticated WebSocket clients and appropriate hub design; use Front Door edge caching with appropriate cache-control and purge policy) address other capabilities instead of this required pair.
Topics fan out a message to independently managed subscriptions with durable handling. The other proposed action (use Front Door edge caching with appropriate cache-control and purge policy) addresses a different operation. Event Grid distributes discrete source events to configured subscribers.
The acceptance criterion for PostgreSQL flexible server comes from this issue: An application depends on PostgreSQL extensions and wants managed PostgreSQL without a VM. Which measured outcome is the strongest proof?
Flexible Server offers managed open-source PostgreSQL with supported extensions and deployment options. Confirm using the measurable result: The required extension is available and the app connects without an OS-managed database host.
Active geo-replication creates readable asynchronously updated secondary databases while primary remains writable. The distinguishing check is: The required extension is available and the app connects without an OS-managed database host.
Partition elimination and aligned indexes can reduce scans when predicates match partition keys. The SQL database partitioning test checks execution plans prune irrelevant historical partitions for the test query; the relevant outcome is The required extension is available and the app connects without an OS-managed database host.
Managed SQL compute tiers can scale resources but scaling may have operational consequences. The offered observation would demonstrate SQL compute scaling: A scheduled scale-up meets throughput and a later scale-down reduces cost. The current validation is different: The required extension is available and the app connects without an OS-managed database host.
During technical sign-off for Delegation with admin units, the architect selects: Create appropriately scoped administrative units and assign suitable delegated Entra administrative roles. Which underlying behavior should be documented?
Administrative units scope eligible directory management roles to selected objects, not data visibility. It is material here because a regional help desk should manage only user accounts in its country without becoming tenant-wide admin.
Key Vault secrets support controlled retrieval and rotation separate from packaged application code. Administrative units scope eligible directory management roles to selected objects, not data visibility.
Azure SQL supports Entra authentication and compatible managed identities can eliminate stored SQL secrets. Administrative units scope eligible directory management roles to selected objects, not data visibility.
Data-plane roles control access to blob content and least-scope assignment reduces unnecessary privileges. It belongs to Managed-identity scope. The stated architecture instead relies on: Administrative units scope eligible directory management roles to selected objects, not data visibility.
A fleet needs immediate warning when sustained processor utilization exceeds its target without reading logs manually. An engineer must deliver the following acceptance evidence: A simulated sustained high-CPU period fires the metric alert and notifies the on-call team. What design choice is defensible?
Metrics are low-latency numeric series and alert evaluation can trigger actions independently of a log query. The selected architecture is create Azure Monitor metric alert rules with appropriate evaluation windows and an action group; confirm the result by checking that a simulated sustained high-CPU period fires the metric alert and notifies the on-call team.
VM insights and the agent collect supported guest operating-system counters and process data. The relevant requirement is: A fleet needs immediate warning when sustained processor utilization exceeds its target without reading logs manually.
Storage-based diagnostic archives and immutability serve compliance preservation rather than interactive KQL search. It does not solve the specified CPU threshold alerts condition.
Container insights exposes Kubernetes workload health and container logs beyond generic virtual-machine statistics. Its design target is AKS container diagnosis. The stated requirement instead concerns: A fleet needs immediate warning when sustained processor utilization exceeds its target without reading logs manually.
Architecture review: SQL schema isolation and SQL elastic pool. Requirement A: A multitenant app needs strong tenant-level database isolation and easy tenant backup/restore. Requirement B: Hundreds of tenant databases have peaks at different times and waste money with dedicated compute. Which complete two-part proposal satisfies both constraints?
For requirement A, database-per-tenant provides a distinct security and backup boundary versus one shared schema. For requirement B, elastic pools share resources across independent databases with non-synchronized demand. Verify both independently: A tenant restore does not overwrite another tenant's data; Aggregated tenant peaks stay within pool limits at lower cost than individually overprovisioned databases.
The option covers one requirement but substitutes SQL database monitoring for SQL elastic pool. The missing acceptance result is: Aggregated tenant peaks stay within pool limits at lower cost than individually overprovisioned databases.
Database-per-tenant provides a distinct security and backup boundary versus one shared schema. Elastic pools share resources across independent databases with non-synchronized demand. The proposed steps (consider Azure SQL Managed Instance after checking required SQL Server features; use Azure SQL query performance insight and Azure Monitor metrics/diagnostics) address other capabilities instead of this required pair.
The first step implements SQL managed instance compatibility instead of SQL schema isolation. The necessary change is to use one managed database per tenant when isolation and lifecycle needs outweigh extra operations.
Architecture rationale review—SQL managed instance compatibility: Consider Azure SQL Managed Instance after checking required SQL Server features. Which explanation accurately distinguishes this capability?
Managed Instance supports many instance-scoped SQL Server capabilities with a managed service. It is material here because a legacy SQL Server app requires SQL Agent jobs and broad instance-level compatibility during cloud migration.
Azure SQL Database offers fully managed relational database operations with automated backups. It belongs to Azure SQL Database single database. The stated architecture instead relies on: Managed Instance supports many instance-scoped SQL Server capabilities with a managed service.
CMK encryption protects database data at rest while keys are managed under customer control. Managed Instance supports many instance-scoped SQL Server capabilities with a managed service.
Flexible Server offers managed open-source PostgreSQL with supported extensions and deployment options. Managed Instance supports many instance-scoped SQL Server capabilities with a managed service.
A solution must address both Application Gateway path routing and Front Door rules engine. First requirement: One regional HTTPS endpoint needs /api requests routed to API pods and /shop to a web tier. Second requirement: An application needs to redirect an old URL path and add a security header at the global edge without changing its origin code. Select TWO recommendations that meet these independent needs.
For the Front Door rules engine half of this design, Front Door rule sets change eligible request and response behavior at the edge according to match conditions. Check: Requests to the deprecated path redirect and supported responses include the configured header. The other half concerns Application Gateway path routing.
For the Application Gateway path routing half of this two-part design, application Gateway provides application-aware path routing rather than basic L4 forwarding. Verify: Requests to each path arrive at the correct healthy backend pool. The other half concerns Front Door rules engine.
Queueing decouples front-end latency from downstream worker availability. Its own validation would be: Checkout responds quickly and pending invoices process after the worker recovers. The current paired requirement is Application Gateway path routing and Front Door rules engine.
Central secret retrieval separates sensitive credentials from source and build artifacts. This alternative solves Key Vault app secrets, while the question separately requires Application Gateway path routing and Front Door rules engine.
APIM applies gateway policies authentication mediation throttling and API lifecycle controls. That plan is appropriate for API Management gateway. The two controls sought here are Application Gateway path routing and Front Door rules engine.
For the Cosmos periodic backup requirement, the proposed architecture uses: Evaluate continuous-backup mode if precise PITR is a firm requirement. Why does that design fit?
Periodic backups produce interval snapshots and are not the same as continuous point-in-time recovery. It is material here because a Cosmos DB workload uses periodic backups and the team expects minute-level arbitrary restores.
RPO measures the maximum tolerated data loss period at a recovery point. Periodic backups produce interval snapshots and are not the same as continuous point-in-time recovery.
Multi-user authorization separates approval boundaries for critical backup changes. It belongs to Multifactor critical operations. The stated architecture instead relies on: Periodic backups produce interval snapshots and are not the same as continuous point-in-time recovery.
Failback is a separate operational sequence after workloads run in the recovery site. That statement describes Failback after recovery, while the question asks how Cosmos periodic backup works: Periodic backups produce interval snapshots and are not the same as continuous point-in-time recovery.
The acceptance criterion for Backup versus DR comes from this issue: Executives expect a nightly VM backup to meet a five-minute regional recovery target. Which measured outcome is the strongest proof?
Nightly backups alone do not provide continuous workload replication or low recovery time. Confirm using the measurable result: The DR drill meets recovery objectives while backup restore still supports historical rollback.
Soft delete retains recoverable protected backup data after deletion for a configured interval. Those observations validate Backup soft delete. They do not establish the required Backup versus DR result: The DR drill meets recovery objectives while backup restore still supports historical rollback.
Periodic backups produce interval snapshots and are not the same as continuous point-in-time recovery. The Cosmos periodic backup test checks the new design demonstrates recovery granularity consistent with the RPO; the relevant outcome is The DR drill meets recovery objectives while backup restore still supports historical rollback.
Azure Backup creates managed recovery points distinct from continuously replicating a running VM. The offered observation would demonstrate Azure VM backup policy: A file-level restore recovers the deleted content from the appropriate recovery point. The current validation is different: The DR drill meets recovery objectives while backup restore still supports historical rollback.
The acceptance criterion for Queue asynchronous processing comes from this issue: Checkout requests should return promptly while invoices are generated asynchronously with retry support. Which measured outcome is the strongest proof?
Queueing decouples front-end latency from downstream worker availability. Confirm using the measurable result: Checkout responds quickly and pending invoices process after the worker recovers.
Front Door rule sets change eligible request and response behavior at the edge according to match conditions. The Front Door rules engine test checks requests to the deprecated path redirect and supported responses include the configured header; the relevant outcome is Checkout responds quickly and pending invoices process after the worker recovers.
Durable orchestrations checkpoint progress and coordinate stateful serverless workflows. The offered observation would demonstrate Durable Functions orchestration: Restarting a worker does not restart already completed business steps. The current validation is different: Checkout responds quickly and pending invoices process after the worker recovers.
Edge content distribution can cache static resources near users. Those observations validate CDN static content. They do not establish the required Queue asynchronous processing result: Checkout responds quickly and pending invoices process after the worker recovers.
Before approving Recovery time objective, the reviewer assesses this requirement: A payment system must resume in thirty minutes after a regional outage. Which of the reported outcomes is decisive?
RTO is the maximum acceptable restoration duration, not the amount of lost data. Confirm using the measurable result: A regional failover test restores business service in less than the allowed interval.
Azure SQL automatic full differential and log backups support restoring database state at a selected time. The SQL database PITR test checks a restored database contains transactions immediately preceding the truncation; the relevant outcome is A regional failover test restores business service in less than the allowed interval.
Backup private connectivity has service and workload prerequisites distinct from ordinary storage private endpoints. Those observations validate Backup private endpoints. They do not establish the required Recovery time objective result: A regional failover test restores business service in less than the allowed interval.
Soft delete retains recoverable protected backup data after deletion for a configured interval. The distinguishing check is: A regional failover test restores business service in less than the allowed interval.
Please check your mailbox for a message from support@examlabs.com and follow the directions.