A would instead validate the other design where a PostgreSQL workload needs recovery to a point before a mistaken transaction, within its retention period. That evidence is consistent with use Azure Database for PostgreSQL Flexible Server point-in-time restore, because managed PostgreSQL backups support restoration to a new server at a supported timestamp. It does not demonstrate that the target case achieves an earlier share snapshot provides a missing test file without replacing the whole account.

B would instead validate the other design where a company needs a recovery window for accidentally deleted objects in its blob container. That evidence is consistent with enable blob soft delete with the required retention, because soft delete retains recoverable deleted blob data during the configured window. It does not demonstrate that the target case achieves an earlier share snapshot provides a missing test file without replacing the whole account.

C would instead validate the other design where a design agency must recover a previously overwritten version of a supported Azure blob. That evidence is consistent with enable blob versioning on the eligible storage account, because versioning preserves prior object versions when a blob is changed. It does not demonstrate that the target case achieves an earlier share snapshot provides a missing test file without replacing the whole account.

“Enable zone redundancy in a supported Azure SQL Database service tier and region” relies on this Azure capability: zone redundant database placement mitigates an entire zonal infrastructure failure. This workload involves A stateless VM-based application must continue serving users if one availability zone fails. The decisive Azure design mechanism is: zonal distribution reduces dependence on one physical failure domain. Thus “Distribute redundant VM instances across availability zones behind an appropriate load balancer” is the applicable architecture choice.

“Choose an eligible zone-redundant Azure App Service plan and adequate instance count” relies on this Azure capability: zone redundancy spreads platform instances across available zones. This workload involves A stateless VM-based application must continue serving users if one availability zone fails. The decisive Azure design mechanism is: zonal distribution reduces dependence on one physical failure domain. Thus “Distribute redundant VM instances across availability zones behind an appropriate load balancer” is the applicable architecture choice.

“Design an AKS deployment using supported multi-zone node pools and scheduling” relies on this Azure capability: zonal node capacity and pod distribution allow surviving nodes to host workload replicas. This workload involves A stateless VM-based application must continue serving users if one availability zone fails. The decisive Azure design mechanism is: zonal distribution reduces dependence on one physical failure domain. Thus “Distribute redundant VM instances across availability zones behind an appropriate load balancer” is the applicable architecture choice.

A would instead validate the other design where a managed web application must keep serving through single-zone datacenter failure. That evidence is consistent with choose an eligible zone-redundant Azure App Service plan and adequate instance count, because zone redundancy spreads platform instances across available zones. It does not demonstrate that the target case achieves forced loss of one zone leaves healthy VM instances receiving traffic.

C would instead validate the other design where a mission-critical Azure SQL database needs protection against single availability-zone loss. That evidence is consistent with enable zone redundancy in a supported Azure SQL Database service tier and region, because zone redundant database placement mitigates an entire zonal infrastructure failure. It does not demonstrate that the target case achieves forced loss of one zone leaves healthy VM instances receiving traffic.

D would instead validate the other design where a containerized microservice workload needs Kubernetes workers resilient to zonal failure. That evidence is consistent with design an AKS deployment using supported multi-zone node pools and scheduling, because zonal node capacity and pod distribution allow surviving nodes to host workload replicas. It does not demonstrate that the target case achieves forced loss of one zone leaves healthy VM instances receiving traffic.

“Use eligible read-access geo-zone-redundant Azure Storage” relies on this Azure capability: zone-redundant primary plus secondary read access supports a defined resilience objective. This workload involves A containerized microservice workload needs Kubernetes workers resilient to zonal failure. The decisive Azure design mechanism is: zonal node capacity and pod distribution allow surviving nodes to host workload replicas. Thus “Design an AKS deployment using supported multi-zone node pools and scheduling” is the applicable architecture choice.

“Put independently deployed regional endpoints behind Azure Front Door health-based routing” relies on this Azure capability: front Door routes eligible HTTP traffic to healthy regional origins. This workload involves A containerized microservice workload needs Kubernetes workers resilient to zonal failure. The decisive Azure design mechanism is: zonal node capacity and pod distribution allow surviving nodes to host workload replicas. Thus “Design an AKS deployment using supported multi-zone node pools and scheduling” is the applicable architecture choice.

“Choose an eligible zone-redundant Azure App Service plan and adequate instance count” relies on this Azure capability: zone redundancy spreads platform instances across available zones. This workload involves A containerized microservice workload needs Kubernetes workers resilient to zonal failure. The decisive Azure design mechanism is: zonal node capacity and pod distribution allow surviving nodes to host workload replicas. Thus “Design an AKS deployment using supported multi-zone node pools and scheduling” is the applicable architecture choice.

A would instead validate the other design where a managed web application must keep serving through single-zone datacenter failure. That evidence is consistent with choose an eligible zone-redundant Azure App Service plan and adequate instance count, because zone redundancy spreads platform instances across available zones. It does not demonstrate that the target case achieves application pods remain routable after one node zone becomes unavailable.

B would instead validate the other design where a public global application must keep operating after the outage of its primary Azure region. That evidence is consistent with put independently deployed regional endpoints behind Azure Front Door health-based routing, because front Door routes eligible HTTP traffic to healthy regional origins. It does not demonstrate that the target case achieves application pods remain routable after one node zone becomes unavailable.

C would instead validate the other design where a media service must tolerate local zone loss and retain a readable secondary-region copy for supported blobs. That evidence is consistent with use eligible read-access geo-zone-redundant Azure Storage, because zone-redundant primary plus secondary read access supports a defined resilience objective. It does not demonstrate that the target case achieves application pods remain routable after one node zone becomes unavailable.

“Put independently deployed regional endpoints behind Azure Front Door health-based routing” relies on this Azure capability: front Door routes eligible HTTP traffic to healthy regional origins. This workload involves A proposed Azure workload has the following constraint: A managed web application must keep serving through single-zone datacenter failure. The decisive Azure design mechanism is: zone redundancy spreads platform instances across available zones. Thus “Choose an eligible zone-redundant Azure App Service plan and adequate instance count” is the applicable architecture choice.

“Use a Virtual Machine Scale Set with redundant instances and appropriate health integration” relies on this Azure capability: a scale set supports coordinated scaling and replacing unhealthy compute instances. This workload involves A proposed Azure workload has the following constraint: A managed web application must keep serving through single-zone datacenter failure. The decisive Azure design mechanism is: zone redundancy spreads platform instances across available zones. Thus “Choose an eligible zone-redundant Azure App Service plan and adequate instance count” is the applicable architecture choice.

“Use Azure Cosmos DB multi-region writes with a suitable conflict-resolution plan” relies on this Azure capability: multiwrite placement reduces dependence on a sole write region. This workload involves A proposed Azure workload has the following constraint: A managed web application must keep serving through single-zone datacenter failure. The decisive Azure design mechanism is: zone redundancy spreads platform instances across available zones. Thus “Choose an eligible zone-redundant Azure App Service plan and adequate instance count” is the applicable architecture choice.

A would instead validate the other design where a global write-intensive Cosmos DB application needs writes accepted by multiple regions where supported. That evidence is consistent with use Azure Cosmos DB multi-region writes with a suitable conflict-resolution plan, because multiwrite placement reduces dependence on a sole write region. It does not demonstrate that the target case achieves requests continue reaching healthy instances after a simulated zone failure.

C would instead validate the other design where a public global application must keep operating after the outage of its primary Azure region. That evidence is consistent with put independently deployed regional endpoints behind Azure Front Door health-based routing, because front Door routes eligible HTTP traffic to healthy regional origins. It does not demonstrate that the target case achieves requests continue reaching healthy instances after a simulated zone failure.

D would instead validate the other design where an asynchronous background service running on Azure VMs needs elastic capacity without one VM being a single point of failure. That evidence is consistent with use a Virtual Machine Scale Set with redundant instances and appropriate health integration, because a scale set supports coordinated scaling and replacing unhealthy compute instances. It does not demonstrate that the target case achieves requests continue reaching healthy instances after a simulated zone failure.

“Deploy a supported SQL Server Always On availability group across redundant Azure VMs” relies on this Azure capability: availability groups replicate databases among guest-managed SQL instances. This workload involves During a cloud design workshop, stakeholders explain: A public global application must keep operating after the outage of its primary Azure region. The decisive Azure design mechanism is: front Door routes eligible HTTP traffic to healthy regional origins. Thus “Put independently deployed regional endpoints behind Azure Front Door health-based routing” is the applicable architecture choice.

“Use a Virtual Machine Scale Set with redundant instances and appropriate health integration” relies on this Azure capability: a scale set supports coordinated scaling and replacing unhealthy compute instances. This workload involves During a cloud design workshop, stakeholders explain: A public global application must keep operating after the outage of its primary Azure region. The decisive Azure design mechanism is: front Door routes eligible HTTP traffic to healthy regional origins. Thus “Put independently deployed regional endpoints behind Azure Front Door health-based routing” is the applicable architecture choice.

“Distribute redundant VM instances across availability zones behind an appropriate load balancer” relies on this Azure capability: zonal distribution reduces dependence on one physical failure domain. This workload involves During a cloud design workshop, stakeholders explain: A public global application must keep operating after the outage of its primary Azure region. The decisive Azure design mechanism is: front Door routes eligible HTTP traffic to healthy regional origins. Thus “Put independently deployed regional endpoints behind Azure Front Door health-based routing” is the applicable architecture choice.

A would instead validate the other design where a SQL Server workload on Azure Virtual Machines requires high availability while retaining guest-level SQL control. That evidence is consistent with deploy a supported SQL Server Always On availability group across redundant Azure VMs, because availability groups replicate databases among guest-managed SQL instances. It does not demonstrate that the target case achieves browser traffic reaches the second healthy region after the first origin fails.

B would instead validate the other design where an asynchronous background service running on Azure VMs needs elastic capacity without one VM being a single point of failure. That evidence is consistent with use a Virtual Machine Scale Set with redundant instances and appropriate health integration, because a scale set supports coordinated scaling and replacing unhealthy compute instances. It does not demonstrate that the target case achieves browser traffic reaches the second healthy region after the first origin fails.

C would instead validate the other design where a stateless VM-based application must continue serving users if one availability zone fails. That evidence is consistent with distribute redundant VM instances across availability zones behind an appropriate load balancer, because zonal distribution reduces dependence on one physical failure domain. It does not demonstrate that the target case achieves browser traffic reaches the second healthy region after the first origin fails.

“Distribute redundant VM instances across availability zones behind an appropriate load balancer” relies on this Azure capability: zonal distribution reduces dependence on one physical failure domain. This workload involves An asynchronous background service running on Azure VMs needs elastic capacity without one VM being a single point of failure. The decisive Azure design mechanism is: a scale set supports coordinated scaling and replacing unhealthy compute instances. Thus “Use a Virtual Machine Scale Set with redundant instances and appropriate health integration” is the applicable architecture choice.

“Design an AKS deployment using supported multi-zone node pools and scheduling” relies on this Azure capability: zonal node capacity and pod distribution allow surviving nodes to host workload replicas. This workload involves An asynchronous background service running on Azure VMs needs elastic capacity without one VM being a single point of failure. The decisive Azure design mechanism is: a scale set supports coordinated scaling and replacing unhealthy compute instances. Thus “Use a Virtual Machine Scale Set with redundant instances and appropriate health integration” is the applicable architecture choice.

“Configure zone-redundant high availability for PostgreSQL Flexible Server” relies on this Azure capability: a warm standby in another availability zone enables managed failover. This workload involves An asynchronous background service running on Azure VMs needs elastic capacity without one VM being a single point of failure. The decisive Azure design mechanism is: a scale set supports coordinated scaling and replacing unhealthy compute instances. Thus “Use a Virtual Machine Scale Set with redundant instances and appropriate health integration” is the applicable architecture choice.

A would instead validate the other design where a stateless VM-based application must continue serving users if one availability zone fails. That evidence is consistent with distribute redundant VM instances across availability zones behind an appropriate load balancer, because zonal distribution reduces dependence on one physical failure domain. It does not demonstrate that the target case achieves unhealthy instances are removed or replaced while background work continues.

C would instead validate the other design where a containerized microservice workload needs Kubernetes workers resilient to zonal failure. That evidence is consistent with design an AKS deployment using supported multi-zone node pools and scheduling, because zonal node capacity and pod distribution allow surviving nodes to host workload replicas. It does not demonstrate that the target case achieves unhealthy instances are removed or replaced while background work continues.

D would instead validate the other design where a PostgreSQL Flexible Server in a supported region must survive a primary availability-zone failure. That evidence is consistent with configure zone-redundant high availability for PostgreSQL Flexible Server, because a warm standby in another availability zone enables managed failover. It does not demonstrate that the target case achieves unhealthy instances are removed or replaced while background work continues.