View Full Microsoft AI-200 Exam Dumps and Practice Test Dumps

 

Question 221

Which Azure AI Search feature configures how vector queries are processed?

1. Synonym map
2. Vector search profile
3. Scoring profile
4. Index alias

Correct Answer: 2

Explanation:

A vector search profile in Azure AI Search defines configuration used for vector search behavior, including the algorithm and related vector-search settings associated with vector fields. This helps applications control how vector similarity is evaluated when retrieving semantically related content. A synonym map handles alternative terms, a scoring profile influences ranking, and an index alias provides an abstraction over a search index. Vector search profiles are important when building AI applications that use embeddings to retrieve documents based on semantic similarity rather than only matching exact words.

Question 222

Which Azure AI Search feature converts text into vectors during indexing?

1. Semantic ranker
2. Search analyzer
3. Integrated vectorization
4. Facet navigation

Correct Answer: 3

Explanation:

Integrated vectorization allows Azure AI Search to incorporate vector generation into supported indexing and querying workflows. Text or other supported content can be transformed into vector representations using configured vectorizers, helping applications build semantic retrieval experiences without manually implementing every embedding step. Semantic ranking improves result ordering, analyzers process textual content, and facets provide categorized navigation. Integrated vectorization is especially useful for retrieval-augmented generation applications where source documents need embeddings before they can participate effectively in vector or hybrid searches.

Question 223

Which Azure Cosmos DB capability executes operations transactionally within one logical partition?

1. Transactional batch
2. Change feed
3. Analytical store
4. Index transformation

Correct Answer: 1

Explanation:

A transactional batch in Azure Cosmos DB allows multiple operations against items in the same logical partition to execute as a single atomic unit. This means the operations can succeed together or fail together, helping applications maintain consistency across related updates. The change feed exposes modifications for downstream processing, analytical store supports analytical workloads, and index transformation relates to indexing behavior. Transactional batches are useful when an AI application must update several related records together, such as maintaining coordinated document metadata and processing-state information.

Question 224

Which Azure Cosmos DB feature enables querying data with SQL syntax?

1. MongoDB API
2. Cassandra API
3. NoSQL API
4. Gremlin API

Correct Answer: 3

Explanation:

Azure Cosmos DB for NoSQL supports SQL-like query syntax for working with JSON documents. Applications can use queries to retrieve and filter document data while taking advantage of Cosmos DB’s distributed architecture. The MongoDB API supports MongoDB-compatible workloads, Cassandra API supports Cassandra-based applications, and Gremlin API is designed for graph workloads. Selecting the appropriate API depends on the application’s data model and compatibility requirements. For document-oriented AI applications, the NoSQL API provides a natural model for storing and querying JSON-based application data.

Question 225

Which Azure Storage feature prevents protected blobs from being modified during retention?

1. Blob versioning
2. Immutable storage
3. Object replication
4. Metadata

Correct Answer: 2

Explanation:

Azure Blob immutable storage can protect data from modification or deletion during a configured retention period. This is useful for workloads that must preserve records in an unaltered state for compliance, auditing, or governance requirements. Blob versioning maintains previous versions, object replication creates copies across supported accounts, and metadata stores application-defined attributes. Immutable storage is especially relevant when AI applications generate records that must remain unchanged for a defined period, such as audit artifacts, regulated documents, or evidence required for compliance processes.

Question 226

Which Azure Storage option enables secure access without exposing account keys?

1. Shared access signature
2. Storage endpoint
3. Access tier
4. Blob metadata

Correct Answer: 1

Explanation:

A shared access signature, or SAS, provides delegated access to Azure Storage resources without requiring applications to distribute the storage account’s primary access keys. A SAS can specify permissions, resources, and validity periods according to its configuration. A storage endpoint identifies a service location, an access tier controls storage characteristics, and blob metadata stores descriptive key-value information. SAS is useful when an AI application needs to grant temporary or limited access to documents, images, datasets, or other stored content without exposing broader account-level credentials.

Question 227

Which Azure Service Bus capability sends several messages as one atomic operation?

1. Message deferral
2. Transaction
3. Subscription filter
4. Peek operation

Correct Answer: 2

Explanation:

Azure Service Bus transactions allow supported messaging operations to participate in an atomic transaction. This can help ensure that related message operations either complete together or are rolled back together when the transaction fails. Message deferral postpones processing, subscription filters determine which topic messages are delivered, and Peek inspects messages without removing them. Transactions are valuable for workflows where consistency between multiple messaging operations matters, such as processing an incoming message while producing related outgoing messages as part of the same logical operation.

Question 228

Which Event Hubs setting determines the number of independent event streams?

1. Consumer group
2. Partition count
3. Capture interval
4. Checkpoint location

Correct Answer: 2

Explanation:

The partition count determines how many partitions an Event Hubs event hub contains, providing separate ordered streams within the hub and enabling parallel event processing. Partitions are an important scalability concept because events are distributed across them according to the configured partitioning behavior. Consumer groups provide independent consumption views rather than creating partitions, Capture controls event-storage behavior, and checkpoint locations store consumer progress. Selecting an appropriate partition structure is important for high-volume AI telemetry and event-processing workloads.

Question 229

Which Event Grid resource can represent an Azure service publishing events?

1. System topic
2. Search index
3. Consumer group
4. Storage queue

Correct Answer: 1

Explanation:

An Event Grid system topic represents events published by an Azure service or supported Azure resource. It provides a managed event source that applications can subscribe to without creating a custom event-publishing mechanism for the Azure resource. A search index belongs to Azure AI Search, a consumer group belongs to Event Hubs, and a storage queue provides asynchronous message storage. System topics are useful in event-driven AI architectures where an application needs to react automatically to events generated by supported Azure resources.

Question 230

Which Azure Functions feature supports workflows containing external events?

1. Durable entities
2. Durable orchestrations
3. Durable external events
4. Function bindings

Correct Answer: 3

Explanation:

Durable Functions external events allow an orchestration to wait for an event raised by an external system or application. This is useful when workflow progress depends on an outside action, such as an approval, callback, or asynchronous result. Durable entities provide stateful coordination, durable orchestrations coordinate workflow steps, and bindings connect functions with supported services. External events can help AI workflows pause safely while waiting for information that is not available immediately, allowing the orchestration to resume when the required event arrives.

Question 231

Which Azure Functions model allows an orchestration to maintain workflow state?

1. Durable orchestration
2. HTTP response
3. Timer schedule
4. Blob trigger

Correct Answer: 1

Explanation:

A Durable Functions orchestration provides a programming model for coordinating stateful, long-running workflows. The Durable Functions runtime manages orchestration state so applications can coordinate multiple activities, timers, retries, and external events without manually implementing all state-persistence infrastructure. HTTP responses and timer schedules represent individual invocation mechanisms, while Blob triggers start functions from blob-related events. Durable orchestrations are particularly useful for AI processing pipelines that contain multiple dependent stages and may need to pause and resume over an extended period.

Question 232

Which Azure Container Apps capability routes traffic between application revisions?

1. Environment variables
2. Traffic splitting
3. Secret references
4. Registry authentication

Correct Answer: 2

Explanation:

Traffic splitting in Azure Container Apps allows incoming application traffic to be distributed among revisions according to configured percentages. This can support controlled rollout strategies where a new application version receives only a portion of traffic before becoming the primary version. Environment variables provide application configuration, secret references expose sensitive values, and registry authentication allows access to private images. Traffic splitting can be useful for AI applications when developers need to test a new service version with limited production traffic before increasing its exposure.

Question 233

Which Azure Container Apps option stores a reusable environment-level configuration value?

1. Environment variable
2. Container replica
3. Ingress endpoint
4. Revision suffix

Correct Answer: 1

Explanation:

Environment variables provide configuration values to containers at runtime. They allow applications to separate configurable behavior from the container image itself, making the same image reusable across multiple environments. Container replicas represent running application instances, ingress endpoints expose services, and revision suffixes help identify deployed versions. Environment variables are useful for AI applications that need environment-specific values such as service endpoints, operational modes, or non-sensitive configuration without rebuilding the container image whenever those values change.

Question 234

Which Azure Monitor capability collects telemetry from supported Kubernetes workloads?

1. Managed Prometheus
2. Action group
3. Workbook
4. Activity Log

Correct Answer: 1

Explanation:

Azure Monitor managed service for Prometheus provides a managed monitoring capability for collecting and analyzing Prometheus metrics from supported Kubernetes environments. It reduces the operational burden of maintaining a separate Prometheus infrastructure while integrating monitoring into Azure’s observability ecosystem. Action groups handle alert notifications and actions, workbooks visualize telemetry, and Activity Log records Azure management operations. Managed Prometheus can be useful for AI workloads running on Kubernetes when application and infrastructure teams already use Prometheus-compatible metrics.

Question 235

Which Azure Monitor component defines where collected data is ingested?

1. Data collection endpoint
2. Metric alert
3. Workbook
4. Availability test

Correct Answer: 1

Explanation:

An Azure Monitor data collection endpoint, or DCE, provides an endpoint for supported Azure Monitor data collection scenarios. It can be used as part of a data collection architecture where telemetry is routed through configured ingestion infrastructure. Metric alerts evaluate telemetry conditions, workbooks display monitoring information, and availability tests check application accessibility. Data collection endpoints are useful when implementing more structured monitoring architectures, particularly where organizations need controlled telemetry ingestion for applications and infrastructure operating across multiple environments.

Question 236

Which Bicep construct creates reusable infrastructure definitions?

1. Parameters
2. Modules
3. Outputs
4. Variables

Correct Answer: 2

Explanation:

Bicep modules allow infrastructure definitions to be separated into reusable components. A module can represent a group of resources or deployment logic that can be called from other Bicep files. This helps organizations avoid duplicating infrastructure definitions across applications and environments. Parameters accept deployment inputs, outputs return values, and variables store reusable expressions within a template. Modules are especially useful for AI platforms that repeatedly deploy common components such as storage, monitoring, networking, or application infrastructure.

Question 237

Which Azure deployment feature previews resource changes before deployment?

1. Deployment history
2. What-if operation
3. Resource lock
4. Activity Log

Correct Answer: 2

Explanation:

The Azure Resource Manager what-if operation previews the changes that a deployment would make before those changes are applied. It can show resources that would be created, modified, or deleted, helping developers validate infrastructure changes before committing them. Deployment history records previous deployment activity, resource locks protect resources from certain changes, and Activity Log records management operations. What-if is particularly useful in infrastructure-as-code workflows because it provides an additional validation step before deploying updates to AI application environments.

Question 238

Which Azure API Management component identifies the actual service receiving API requests?

1. Product
2. Backend
3. Subscription
4. Developer portal

Correct Answer: 2

Explanation:

An API Management backend represents the service to which APIM forwards API requests after gateway processing. Backends can point to application services or other supported endpoints that provide the actual API functionality. Products group APIs for consumers, subscriptions provide consumer access arrangements, and the developer portal presents API documentation and management features to developers. Backends are important in AI API architectures because they allow APIM to act as a managed gateway between external clients and the services performing model inference, retrieval, data processing, or other application functions.

Question 239

Which Azure Database for PostgreSQL option provides connection pooling?

1. PgBouncer
2. Query Store
3. Elastic pool
4. Geo-fence

Correct Answer: 1

Explanation:

PgBouncer is a lightweight connection-pooling component supported with Azure Database for PostgreSQL Flexible Server scenarios. Connection pooling can reduce the overhead associated with repeatedly creating and closing database connections, which is particularly useful for applications that create many short-lived connections. Query Store focuses on query performance information, elastic pools are associated with Azure SQL Database, and geo-fencing is unrelated to PostgreSQL connection pooling. For AI applications with variable workloads, connection pooling can help improve database connection efficiency and reduce unnecessary connection-management overhead.

Question 240

Which Azure governance service groups subscriptions into an administrative hierarchy?

1. Resource tags
2. Management groups
3. Resource locks
4. Policy definitions

Correct Answer: 2

Explanation:

Azure management groups organize subscriptions into a hierarchical structure, allowing governance and policy controls to be applied across multiple subscriptions. This is useful for organizations that operate separate environments, business units, or application subscriptions while needing centralized governance. Resource tags add metadata to resources, resource locks protect resources from certain operations, and policy definitions describe governance rules. Management groups can help large AI environments maintain consistent governance across many Azure subscriptions while still allowing individual teams to manage their application resources.