1z0-1147 Oracle AI Database@Azure Architect Professional Questions and Answers

ZDM manages migrations as persistent scheduled jobs and assigns each job a unique job ID. Administrators must manipulate these jobs through the supported ZDMCLI lifecycle commands rather than modifying ZDM repository or log files manually.

Oracle documents the command zdmcli abort job -jobid < job_id > for terminating a specified ZDM job. When an existing job context is preventing a new migration workflow from being established for the same source, the supported administrative approach among the available options is to identify the existing job and abort it through ZDMCLI.

Restarting the ZDM service does not remove the persistent job record; the job metadata is maintained by ZDM rather than existing only in the running daemon's memory. Manually deleting files from ZDM directories is unsupported and risks corrupting or desynchronizing job state. Likewise, the proposed -ignore ALL option is not a supported mechanism for bypassing ZDM's migration-job lifecycle controls.

Consequently, B is the correct operational command because it acts on the existing migration through the supported ZDM control interface.

Study Guide reference: Migrate Oracle Databases to Oracle AI Database@Azure — ZDMCLI job management, query/abort operations and migration workflow lifecycle.

===============

Oracle explicitly requires Advanced Networking for compatibility between Oracle AI Database@Azure and certain Azure services that use specialized private-networking behavior. Oracle's troubleshooting documentation specifically identifies Azure Container Apps and serverless services such as Azure Functions using private endpoints as examples.

Without Advanced Networking, these services may be unable to establish supported connectivity to the Oracle AI Database@Azure delegated subnet even when basic VNet routing appears correct. The feature enables the enhanced networking capabilities required for these communication patterns and is also relevant to Global VNet Peering scenarios.

Oracle notes an important provisioning condition: Advanced Networking should be enabled before Oracle AI Database@Azure resources/VNICs have been provisioned in the delegated subnet for the feature to be applied correctly to subsequent resources.

An NVA route bypass does not provide this compatibility feature. FastConnect is an OCI dedicated-connectivity technology, and ExpressRoute Global Reach addresses ExpressRoute interconnection scenarios rather than the specific Azure Container Apps compatibility requirement.

Thus, C. Advanced Networking is the documented solution.

Study Guide reference: Configure Oracle AI Database@Azure Networking — Advanced Networking, delegated subnet connectivity, Azure Container Apps and serverless/private endpoint integration.

===============

The Network Anchor abstraction establishes the mapping between Azure networking resources and corresponding OCI networking constructs. It is particularly important for services that use the newer multicloud anchor model, allowing an Azure VNet and delegated subnet to be associated consistently with Oracle-managed resources.

Oracle's DNS documentation explicitly restricts the Network Anchor DNS-resolution capability in question to Oracle Base Database Service . Network Anchors can provide DNS integration between Azure and OCI, including inbound and outbound resolution paths. Depending on the configuration, DNS listening and forwarding functionality can be enabled so resources on either side of the multicloud boundary can resolve the required private names.

The question's "Replicate DNS Private zones during DB creation" behavior belongs to this Base Database Service Network Anchor DNS workflow. Exadata Database Service on Dedicated Infrastructure and Exadata Database Service on Exascale Infrastructure use their documented Exadata networking and DNS provisioning mechanisms. Autonomous AI Database likewise has its own networking lifecycle and is not the service to which this Network Anchor-specific private-zone replication setting applies.

Therefore, the database service required to expose the specified option is Oracle Base Database Service .

Study Guide reference: Configure Oracle AI Database@Azure Networking — Network Anchors, private DNS, inbound/outbound DNS resolution, Base Database Service networking.

===============

When Oracle Autonomous AI Database Serverless is provisioned through Oracle AI Database@Azure, the administrative database account follows the standard Autonomous Database account model. The Azure creation workflow exposes the Username field as read-only and set to ADMIN .

The administrator therefore supplies the password for the predefined ADMIN account but does not choose or rename the account during database creation. Oracle currently specifies password requirements separately, including minimum and maximum length and character-composition requirements.

The account is not derived from the Microsoft Entra ID identity that deploys the Azure resource. Azure RBAC and Microsoft Entra authentication govern access to the Azure resource-management plane, whereas the ADMIN credential is a database-level account. Similarly, SYS is not presented as the configurable provisioning username for Autonomous AI Database Serverless, and the username is not generated from subscription metadata.

This separation between Azure control-plane identity and Oracle database credentials is architecturally important when designing authentication and administrative access.

Therefore, the correct answer is A .

Study Guide reference: Provision and operate Oracle AI Database@Azure Resources — Autonomous AI Database Serverless creation, configuration and administrator credentials.

===============

Oracle Application Continuity (AC) is the application-availability technology designed to mask recoverable database outages and replay database requests transparently after a connection is re-established.

Oracle RAC provides database-service continuity when one database instance or node becomes unavailable, but an application may still have active sessions or in-flight work on the failed instance. Application Continuity works with Oracle's high-availability connection infrastructure, database services, FAN, Transaction Guard, and supported clients to reconnect to an available instance and replay eligible database requests. This significantly reduces the application errors that would otherwise be exposed during node failure or planned maintenance.

Fast-Start Failover and Data Guard Broker operate at the Data Guard database-role and disaster-recovery level. They manage primary/standby transitions rather than transparently replaying an individual application session after a RAC node failure. Transparent Data Encryption protects database data at rest and has no role in request replay or failure masking.

Therefore, C. Oracle Application Continuity is correct.

Study Guide reference: HA and DR with Oracle AI Database@Azure — RAC application availability, FAN, Transaction Guard and Application Continuity.

===============

The Data Guard Observer is an independent process used by Fast-Start Failover to monitor the primary and standby databases and determine whether an automatic failover should occur. Because its independence directly affects the reliability of the failover decision, it should not share the same failure domain as either database.

For an Oracle AI Database@Azure architecture spanning availability zones, placing the Observer on a separate Azure VM in another availability zone provides appropriate fault isolation. If the primary AZ fails, the Observer remains available to verify the failure and initiate the transition to the standby. Hosting the Observer directly on the primary database node would cause it to disappear with a primary-site outage. Hosting it on the standby similarly creates an undesirable dependency on the standby environment.

The Observer is an external Data Guard Broker process; it does not belong inside Oracle's multicloud control-plane Parent Site. For stronger resilience, Oracle Data Guard can also support multiple observers in appropriate configurations.

Therefore, D represents the recommended architecture.

Study Guide reference: HA and DR with Oracle AI Database@Azure — Data Guard Broker, FSFO and Observer placement.

===============

Oracle Exadata VM Cluster compute resources can be scaled down to zero OCPUs when the objective is to suspend compute consumption without deleting the VM Cluster definition or its persistent database storage. Oracle explicitly notes that simply stopping VMs does not stop the relevant compute billing; to stop VM Cluster compute-related billing, the enabled OCPU—or ECPU on applicable generations—allocation must be reduced to zero.

At zero enabled compute, the guest VMs are shut down while the infrastructure configuration and persistent database storage remain available. The administrator can later restore service by scaling the enabled compute allocation back above zero. This makes the feature useful for development, testing, seasonal, or otherwise intermittent environments where deleting and recreating the database environment would be operationally undesirable.

Values of one, two, or four OCPUs still represent allocated active compute and therefore do not achieve the requested zero-compute billing state.

On Exascale Infrastructure, Oracle documents the equivalent ECPU behavior explicitly: enabled ECPUs can be scaled to zero, shutting down VM Cluster VMs and stopping enabled-ECPU usage billing.

Study Guide reference: Provision and operate Oracle AI Database@Azure Resources — VM Cluster scaling, OCPU/ECPU allocation and cost control.

===============

Under the default managed-DNS model, Oracle derives the private Oracle domain from normalized VNet and delegated-subnet naming components. In the source scenario, the production and DR resource names share the same leading characters, causing both sites to derive the same oraclevcn.com private DNS namespace.

The practical resolution is therefore to alter the naming inputs before deployment so the generated DNS labels are unique. The uploaded question set specifies that the relevant generated tokens use the first 11 alphanumeric characters , so either the DR VNet or delegated subnet must be renamed such that the derived prefix differs from the active environment.

The Host Domain Name field associated with default managed DNS is system-generated rather than freely editable. Adding another DNS resolver does not eliminate the underlying duplicate private-zone namespace, and DNS views cannot reliably correct a collision created at resource provisioning.

Therefore, D resolves the root cause while retaining Oracle-managed DNS.

Study Guide reference: Configure Oracle AI Database@Azure Networking — default DNS, oraclevcn.com, VNet/subnet DNS derivation and DR naming design.

===============

Oracle's multicloud guidance identifies business and architecture patterns rather than narrowly defined infrastructure features as core multicloud use cases. Two explicitly documented patterns are Cross-Cloud Disaster Recovery and Running Applications in One Cloud and Databases in Another .

Cross-cloud disaster recovery places a production environment in one cloud while maintaining a recovery environment in another. This design reduces dependence on a single cloud platform and enables business continuity if the primary environment experiences a major outage. It is particularly relevant where resilience requirements justify provider-level diversification.

The application/database split pattern allows an organization to operate its application tier in one cloud while placing its database in another cloud that provides preferable database capabilities, performance, economics, or management characteristics. Oracle AI Database@Azure is a practical implementation of this model: Azure applications can use Oracle database technology deployed in Azure data centers and managed through Oracle's multicloud architecture.

Automatic identity-domain replication and Active Directory domain-controller synchronization may support a multicloud environment, but they are implementation capabilities rather than the high-level multicloud use cases documented by Oracle.

Study Guide reference: Introduction to Multicloud — multicloud drivers, application/database separation, cross-cloud DR, resiliency and workload placement.

===============

Oracle's cross-region Oracle Database@Azure reference architecture requires an additional Hub VCN because each OCI VCN can be attached to only one Dynamic Routing Gateway (DRG) at a time.

The Oracle-managed database VCN associated with an Exadata VM Cluster already uses a DRG as part of the Oracle Database@Azure network architecture. Cross-region VCN peering also requires DRG-based remote peering. Because a second DRG cannot simply be attached to the existing database VCN, Oracle introduces a separate Hub VCN with its own DRG. The database VCN is locally connected to the Hub VCN, and the Hub DRGs establish the cross-region remote peering connection.

OCI documentation independently confirms that one VCN can attach to only one DRG, whereas a single DRG can support many VCN attachments. A VCN can also have multiple Local Peering Gateways, so option B is technically incorrect.

The limitation is therefore architectural rather than a restriction caused by Azure subnet delegation or tenancy boundaries.

Study Guide reference: HA and DR — OCI transit networking, Hub VCN, DRG, LPG and cross-region Data Guard.

===============

The -eval option invokes the evaluation workflow of Oracle Zero Downtime Migration rather than executing the production migration. Oracle specifically directs administrators to use ZDMCLI migrate database ... -eval to evaluate the migration process, including prerequisite checks, source/target connectivity verification, configuration validation, and applicable pre-migration assessment phases.

The resulting job is an evaluation job rather than a migration job that performs database movement. This distinction is reflected throughout ZDM command handling: ZDM can query evaluation jobs separately from actual migration jobs, and the evaluation workflow is intended to identify prerequisite or configuration failures before production data movement begins.

The option does not create a mock standby or execute a simulated Data Guard switchover. It also does not begin an offline migration and stop before the copy phase. CPAT may participate in migration-readiness assessment depending on ZDM version and command options, but -eval is not an installation command for CPAT.

Consequently, A correctly represents the effect: an EVAL-type workflow performs migration prechecks and validation without executing the actual migration.

Study Guide reference: Migrate Oracle Databases to Oracle AI Database@Azure — ZDM CLI, evaluation mode, migration validation and prechecks.

===============

OCI Vault is Oracle Cloud Infrastructure's centralized service for managing encryption keys and secrets. In Oracle AI Database@Azure, it can participate in customer-managed key architectures for Oracle Transparent Data Encryption.

TDE encrypts Oracle database data at rest using master encryption keys. Oracle AI Database@Azure supports Oracle-managed keys stored in an Oracle Wallet as the default model and several customer-managed key stores, including OCI Vault, Oracle Key Vault, and Azure Key Vault. Using OCI Vault allows organizations to manage cryptographic key lifecycle operations—including controlled creation, rotation, access policy, and auditing—outside the database itself.

Vault and identity management should not be confused. Microsoft Entra ID and OCI IAM Identity Domains implement identity federation and user authentication; OCI Vault does not function as an identity provider. Likewise, it is unrelated to content delivery or application/network connectivity. Delegated subnets remain mandatory Azure networking constructs regardless of the selected encryption-key architecture.

Therefore, the correct description is A .

Study Guide reference: Secure Oracle AI Database@Azure — TDE, Oracle-managed and customer-managed encryption keys, OCI Vault.

===============

Oracle AI Database@Azure depends on Azure Resource Manager provider namespaces that expose the Oracle service and its underlying infrastructure integration. Among the listed choices, the required providers are Microsoft.BareMetal and Oracle.Database .

Oracle.Database is the Azure resource-provider namespace through which Oracle database multicloud resources are represented and managed in Azure. This includes Oracle AI Database@Azure database-resource lifecycle operations and integration with delegated networking. Microsoft.BareMetal supports the Azure-side integration with specialized Oracle infrastructure physically deployed within Azure facilities.

Current Microsoft onboarding documentation also lists Microsoft.Network and Microsoft.Compute among resource providers required for a fully provisionable subscription. However, those names are not offered as valid alternatives in this question. Microsoft.VirtualMachine is not the correct provider namespace—standard Azure compute uses Microsoft.Compute. Likewise, Oracle.Exadata is not the service namespace; Oracle AI Database@Azure resources are represented under Oracle.Database.

Therefore, among the supplied answer set, B and D are correct.

Study Guide reference: Oracle AI Database@Azure Architecture and Onboarding — Azure subscription prerequisites, resource-provider registration and Oracle.Database integration.

===============

The scenario directly matches Oracle's documented multicloud use case of running applications in one cloud and databases in another , also described here as an application and database split-stack architecture .

This design allows an organization to place its application tier in the cloud offering the preferred application-development ecosystem while retaining the database on the cloud platform offering the required database capabilities. In this case, Azure provides containerization and application modernization services, while OCI continues hosting Oracle Exadata Database Service for high transaction throughput, database availability, and Oracle-native operational control.

Private, high-bandwidth intercloud connectivity is critical because application-to-database latency directly affects transaction response time. Keeping data movement predictable and minimizing unnecessary public-internet routing also improves security and can reduce data-transfer overhead.

Horizontal workload distribution instead spreads workloads across clouds rather than separating application and database tiers. Production/development separation addresses lifecycle environments, while SaaS/ERP integration addresses data interoperability rather than a transactional split-stack application.

Therefore, C is the correct architectural pattern.

Study Guide reference: Introduction to Multicloud — Running Applications in One Cloud and Databases in Another.

===============

Oracle's reference architecture for multi-region Oracle Database@Azure disaster recovery explicitly states that no subnet is required in the Hub VCN to enable transit routing .

The Hub VCN serves primarily as a routing construct connecting gateways and route-table attachments. DRGs, Local Peering Gateways, Remote Peering Connections, and VCN route tables provide the required network forwarding between the Oracle Database@Azure Child Site VCNs and remote regions. Because the hub does not need to host compute instances or network virtual appliances for this gateway-based transit model, it does not require a workload subnet.

This has an additional design benefit: the hub VCN can use a comparatively small CIDR range because there is no requirement to reserve substantial address space for servers or VNICs.

A public subnet and Internet Gateway are unnecessary because Data Guard traffic should traverse private cloud networking. Likewise, dedicated /24 transit-appliance subnets or per-availability-domain private subnets are not mandatory for the gateway-based topology.

Therefore, D is correct.

Study Guide reference: HA and DR with Oracle AI Database@Azure — cross-region Data Guard, OCI Hub VCN and transit routing.

===============

Oracle owns and operates the underlying Exadata infrastructure layer used by Oracle AI Database@Azure, even though that infrastructure is physically located in Microsoft Azure data centers. Consequently, Oracle performs infrastructure-level maintenance, patching, security updates, monitoring, and lifecycle operations.

Oracle-managed Exadata infrastructure components include the physical database-server hosts, storage servers, network fabric, hypervisor or virtualization infrastructure, firmware, management components, and control-plane systems. Microsoft supplies the Azure data-center environment and associated Azure platform capabilities, but Microsoft support engineers do not maintain Oracle's Exadata hardware/software stack.

The customer retains responsibility for customer-controlled portions of co-managed database environments, including database configuration and, depending on service and maintenance model, guest VM, Grid Infrastructure, database homes, and database software lifecycle activities. Oracle has also introduced optional Oracle-managed database software update capabilities for applicable newer Exadata services, but that does not change responsibility for the underlying infrastructure.

Therefore, C correctly identifies Oracle's Cloud Infrastructure operations organization.

Study Guide reference: Provision and operate Oracle AI Database@Azure Resources — shared operational responsibility and Exadata infrastructure maintenance.

===============

Oracle Autonomous AI Database Serverless allows administrators to configure the automatic backup-retention window according to business recovery and compliance requirements. Oracle's current Autonomous AI Database provisioning documentation specifies an allowable automatic backup-retention range of 1 through 60 days .

Within that period, the database can be restored and recovered to an appropriate point in time, subject to the normal Autonomous AI Database backup and recovery mechanisms. A longer retention window provides a broader point-in-time recovery horizon but must also be considered alongside governance, compliance, and backup-protection requirements.

Oracle also supports Backup Retention Lock , which materially changes administrative behavior. When the retention period is locked, administrators cannot simply disable the lock or alter the protected retention setting through ordinary provisioning controls; Oracle Support intervention is required for those changes. This distinction is important for regulated workloads where backup immutability or protection from malicious administrative actions is required.

The alternatives beginning at seven days incorrectly raise the minimum, while the 30-day options incorrectly reduce the supported maximum. The complete configurable range is therefore 1–60 days .

Study Guide reference: Secure Oracle AI Database@Azure — Autonomous AI Database automatic backups, point-in-time recovery, backup-retention configuration and retention lock.

Oracle's current Oracle AI Database@Azure service matrix explicitly identifies Oracle Exadata Database Service on Exascale Infrastructure as eligible for both Private Offer and Public Offer purchasing models.

A Public Offer uses a pay-as-you-go Azure Marketplace model and allows customers to begin consuming supported services without first negotiating a custom contract with Oracle Sales. A Private Offer is based on negotiated commercial terms and is appropriate for enterprise commitments and customized pricing.

Exadata Database Service on Dedicated Infrastructure is Private Offer eligible but not Public Offer eligible. The same Private-Offer-only restriction applies to Autonomous AI Database on Dedicated Exadata Infrastructure. Oracle Database Autonomous Recovery Service is a database protection service rather than the alternative primary high-performance Exadata database platform requested by the scenario, and it is also listed as Private Offer only.

Exascale therefore combines Exadata database capabilities with a lower-entry, elastic shared-infrastructure model and marketplace flexibility.

Thus, A is correct.

Study Guide reference: Oracle AI Database@Azure Architecture and Onboarding — Azure Marketplace offers and available database services.

===============

Oracle Maximum Availability Architecture Silver adds Oracle Real Application Clusters (RAC) to the MAA Bronze foundation. RAC provides active-active database clustering across multiple database-server nodes while all instances access the same underlying database storage.

For Exadata Database Service on Dedicated Infrastructure, RAC supplies localized protection against database-instance and database-server failure. If one RAC instance or compute node fails, services and connections can relocate to surviving instances that are already running. RAC also materially reduces or eliminates database downtime for many planned maintenance operations.

Oracle Active Data Guard addresses a different availability layer: it maintains a physically separate standby database and is primarily used for site, availability-zone, or regional disaster recovery. Oracle GoldenGate performs logical data replication and integration, while Oracle Sharding distributes data across independent database shards for scale and fault isolation.

Oracle's current MAA documentation explicitly identifies RAC as the technology introduced at the Silver tier and recommends Exadata as the optimized RAC platform.

Therefore, B is correct.

Study Guide reference: HA and DR with Oracle AI Database@Azure — MAA Silver, Oracle RAC and local high availability.

===============

With the Default DNS option, Oracle automatically creates and manages the private DNS configuration associated with the Exadata VM Cluster. The generated database networking domain uses the oraclevcn.com suffix.

The Single Client Access Name (SCAN) listener and database-node FQDNs are registered within this Oracle-managed private DNS domain. OCI hosts the corresponding private DNS zone and, as part of the Oracle AI Database@Azure integration, the service replicates the required DNS records into a matching Azure Private DNS zone. Azure then links that private zone to the VNet hosting the Oracle database deployment, allowing Azure applications to resolve the Oracle database endpoints privately.

oraclecloud.com is broadly used for Oracle public cloud endpoints but is not the default private VCN suffix for the Exadata SCAN name. azure-database.net and oci-exadata.com are likewise not the default suffixes used by this Oracle-managed DNS architecture.

Therefore, D. oraclevcn.com is correct.

Study Guide reference: Configure Oracle AI Database@Azure Networking — Exadata default DNS, SCAN and Azure Private DNS replication.

===============

The native architecture is to configure Azure Monitor diagnostic settings on the Oracle Exadata resources and route the selected log categories into an Azure Log Analytics workspace .

Oracle AI Database@Azure exposes relevant Exadata lifecycle and operational events through Azure diagnostic settings without requiring agents to be installed on Oracle-managed Exadata nodes. Once the telemetry reaches Log Analytics, administrators can query it using Kusto Query Language (KQL) and correlate the Oracle resource events with telemetry generated by Azure application infrastructure.

This architecture satisfies the requirement for centralized Azure-side observability while preserving the managed-service boundary of Exadata. It also integrates naturally with Azure dashboards, alerting, Microsoft Sentinel, and other Azure Monitor capabilities.

Using OCI Object Storage would introduce an unnecessary cross-platform telemetry pipeline. Oracle Enterprise Manager remains useful in broader Oracle management scenarios but is not required for this native Azure integration. Installing Azure Monitor Agent directly on Oracle-managed Exadata infrastructure is neither necessary nor the documented implementation pattern.

Therefore, A provides the supported integrated monitoring architecture.

Study Guide reference: Provision and operate Oracle AI Database@Azure Resources — Azure Monitor, diagnostic settings, Log Analytics, KQL and Exadata observability.

===============

The uploaded study set keys D , because it attempts to preserve data inside the same geographical region while obtaining a separate fault domain.

There is, however, an important current-documentation qualification. Oracle AI Database@Azure maps an OCI Availability Domain to an Azure Availability Zone , and Oracle MAA documentation describes local Data Guard standby deployments on another Exadata infrastructure in the same or a separate Azure Availability Zone . Oracle's cross-zone Data Guard implementation explicitly requires primary and standby Oracle AI Database@Azure deployments in two supported Azure AZs.

Therefore, option D is valid only if the second OCI Availability Domain corresponds to another supported Oracle AI Database@Azure Azure AZ. If the premise is literally that Oracle AI Database@Azure is available in only one Azure AZ and cross-region replication is prohibited, current Oracle documentation does not provide a second independent Oracle AI Database@Azure site simply by selecting an unmapped OCI AD.

For the uploaded question's intended key, the answer remains D , but the architecture should be treated with this deployment-availability condition.

===============

Oracle's Multicloud Architecture Framework emphasizes architectural considerations that arise because workloads, applications, data, identities, and operational responsibilities span multiple cloud providers. Two major areas are network latency/data movement and security, identity, and operations management .

Cross-cloud network latency can directly affect application response time, database performance, replication behavior, and transaction processing. Data movement can also introduce throughput constraints and cost considerations. Oracle therefore recommends workload placement and private connectivity strategies that minimize unnecessary cross-cloud distance and traffic.

Security and operations become similarly important because each cloud provider has its own IAM, monitoring, automation, and administrative models. Oracle recommends consistent identity and access policies, unified visibility, standardized automation, and common operational practices to avoid fragmented governance or management silos.

Hardware manufacturing origin and OCI hypervisor patch schedules do not represent the principal multicloud architectural design considerations described by the framework. Hypervisor maintenance is primarily a provider operational responsibility.

Therefore, B and D align directly with Oracle's documented multicloud design pillars.

Study Guide reference: Introduction to Multicloud — latency, data movement, Cloud Security, Cloud Operations, identity integration and operational standardization.

===============

Oracle documents a specific 404 NotAuthorizedOrNotFound condition that can occur during Exadata VM Cluster creation when the selected private DNS zone does not already contain tags. Oracle then attempts to generate the default OCI oracle-tags, but the multicloud service principal must have permission to use the associated tag namespace.

The required policy statement is:

Allow any user to use tag-namespaces in tenancy where request.principal.type = 'multicloudlink'

Oracle's current troubleshooting guidance also requires the companion permission:

Allow any user to manage tag-defaults in tenancy where request.principal.type = 'multicloudlink'

The crucial detail is that the provisioning operation executes under the multicloudlink principal type , not an ordinary administrator group or a principal with request.user.id='multicloud'. Merely granting inspect permission is insufficient because the service must actually apply the tag namespace/defaults.

Thus, among the supplied options, C is the required policy statement.

Study Guide reference: Secure Oracle AI Database@Azure — OCI IAM, MulticloudLink service principal and tagging permissions.

===============

The delegated subnet is the Azure networking construct that enables Oracle-managed database resources to connect directly into the customer's Azure Virtual Network.

During Oracle AI Database@Azure provisioning, a subnet in the customer's VNet is delegated to the Oracle database resource provider. This delegation permits Oracle-managed networking components and database VNICs to participate in the Azure network while preserving the managed-service boundary between customer-owned Azure networking and Oracle-operated database infrastructure.

Applications can then connect privately and with low latency to the Oracle database through addresses associated with that delegated subnet. The customer can also apply supported Azure routing and security constructs around the VNet design.

A Local Peering Gateway connects OCI VCNs within an OCI region. A Dynamic Routing Gateway provides OCI routing between VCNs, on-premises environments, or remote networks. A Service Gateway provides private connectivity from an OCI VCN to supported Oracle services. None of these is the Azure-side resource that places Oracle database networking into the customer's VNet.

Thus, C. Delegated subnet is correct.

Study Guide reference: Configure Oracle AI Database@Azure Networking — delegated subnet architecture and Azure VNet connectivity.

===============

Oracle AI Database@Azure applies a regional service limit to delegated subnet usage. The documented default is five delegated subnets per OCI tenancy in a specific region . This is a service quota rather than a limitation that can be bypassed through additional Azure routing constructs.

To exceed the default, the customer must request a service limit increase . Oracle identifies the relevant OCI service as Multicloud, the resource as Delegated Subnet Multicloud Links , and the corresponding service-limit name as azure-delegated-subnet-count. The limit is regional, so introducing another VNet, Route Server, VNet peering relationship, or Microsoft Entra tenant does not increase the permitted count.

Current documentation phrases the operation as requesting a service limit increase through Oracle's service-limit process. Among the supplied choices, option D is the only one that represents that required Oracle-side quota increase.

Architecturally, this distinction matters because routing topology and service capacity are independent control planes: networking objects determine reachability, whereas Oracle-enforced service limits determine how many delegated-subnet links may exist.

Study Guide reference: Configure Oracle AI Database@Azure Networking — delegated subnet limits and Multicloud service-limit management.

===============

Oracle's current service-availability matrix states that Oracle Exadata Database Service on Dedicated Infrastructure is eligible for Private Offer only . It is not available through the Public Offer/Pay As You Go purchasing model.

Therefore, if an organization originally onboarded Oracle AI Database@Azure using the Public Offer and subsequently requires Dedicated Exadata Infrastructure, the commercial relationship must be changed by engaging Oracle Sales and obtaining the appropriate Private Offer .

A delegated-subnet service-limit increase affects networking capacity and does not enable a database service that is excluded by the selected Marketplace offer. BYOL can reduce Oracle Database licensing charges where eligible, but licensing model and Marketplace service eligibility are separate issues. Changing an Autonomous Database workload type is completely unrelated to Exadata hardware procurement.

Oracle's current service table also shows why the confusion can arise: Exadata Database Service on Exascale Infrastructure is available through both Public and Private Offers, whereas Exadata Database Service on Dedicated Infrastructure remains Private Offer only.

Therefore, A is correct.

Study Guide reference: Oracle AI Database@Azure Architecture and Onboarding — offer eligibility and Dedicated Infrastructure purchasing.

===============