A Region selection determines geography but does not automatically make an application multi-Region; that caveat is associated more closely with AWS Region for the need to select a geographic area in which to place a workload than with AWS Global Accelerator for the requirement to route user traffic toward healthy regional endpoints through AWS global network paths.

An edge site is not a full substitute for application compute in an AWS Region; that caveat is associated more closely with Edge location for the need to deliver cached content closer to geographically dispersed end users than with AWS Global Accelerator for the requirement to route user traffic toward healthy regional endpoints through AWS global network paths.

Multiple AZs help with facility-level failures without being geographically separate Regions; that caveat is associated more closely with Availability Zone for the need to separate redundant application components across isolated facilities within one Region than with AWS Global Accelerator for the requirement to route user traffic toward healthy regional endpoints through AWS global network paths.

AWS SDKs primarily supports the earlier need to integrate AWS service requests directly into application source code because sDKs provide language libraries and request handling for application integrations with AWS APIs, but the new task is to interact programmatically with AWS using supported service requests; that is why AWS service APIs is preferable here.

AWS Systems Manager is intended to manage operational tasks across supported compute fleets because systems Manager provides inventory, automation, patching and remote operational capabilities; in this design the new outcome is to interact programmatically with AWS using supported service requests, which calls for AWS service APIs rather than AWS Systems Manager or the earlier AWS SDKs.

AWS Service Catalog is intended to offer approved infrastructure products through governed self-service because service Catalog lets administrators define centrally governed products users can launch; in this design the new outcome is to interact programmatically with AWS using supported service requests, which calls for AWS service APIs rather than AWS Service Catalog or the earlier AWS SDKs.

Amazon Comprehend addresses the separate need to extract sentiment entities and other insights from text because comprehend provides managed natural-language processing APIs; it is not as direct as Amazon Transcribe for the actual requirement to convert spoken audio into a text transcript.

Amazon Q addresses the separate need to offer generative AI assistance for supported business and developer workflows because Amazon Q provides AI-powered assistant capabilities in integrated AWS settings; it is not as direct as Amazon Transcribe for the actual requirement to convert spoken audio into a text transcript.

Amazon Translate addresses the separate need to translate text between supported human languages because translate is a machine-translation service used by multilingual applications; it is not as direct as Amazon Transcribe for the actual requirement to convert spoken audio into a text transcript.

Amazon Route 53 addresses the separate need to resolve application domain names and apply supported DNS routing policies because Route 53 is a managed DNS and domain service with traffic-routing and health-check features; it is not as direct as Public subnet for the actual requirement to place resources in a subnet with a route toward an internet gateway.

AWS Direct Connect addresses the separate need to establish dedicated private network connectivity from a site to AWS because direct Connect provides a dedicated network connection through supported locations; it is not as direct as Public subnet for the actual requirement to place resources in a subnet with a route toward an internet gateway.

Private subnet addresses the separate need to place backend resources without a direct internet gateway route because a private subnet can reach only the networks available through its route table; it is not as direct as Public subnet for the actual requirement to place resources in a subnet with a route toward an internet gateway.

Amazon API Gateway addresses the separate need to expose and manage HTTP REST or WebSocket APIs because aPI Gateway handles API front doors with request routing and supported authorization options; it is not as direct as Public subnet for the actual requirement to place resources in a subnet with a route toward an internet gateway.

Vertical EC2 scaling addresses the separate need to increase resource capacity by changing the instance type because vertical scaling provides more CPU or RAM within a single virtual machine; it is not as direct as General purpose EC2 instance for the actual requirement to balance CPU memory and networking for broadly mixed workloads.

AWS Elastic Beanstalk addresses the separate need to deploy supported applications while the platform provisions common resources because elastic Beanstalk simplifies web-app deployment and orchestrates the supporting AWS components; it is not as direct as General purpose EC2 instance for the actual requirement to balance CPU memory and networking for broadly mixed workloads.

Gateway Load Balancer addresses the separate need to insert compatible virtual network appliances into traffic paths because gateway Load Balancer distributes flows across appliance targets for inspection or similar tasks; it is not as direct as General purpose EC2 instance for the actual requirement to balance CPU memory and networking for broadly mixed workloads.

EC2 instance store primarily supports the earlier need to use temporary host-attached storage with very low latency because instance store volumes are physically associated with the host and have ephemeral characteristics, but the new task is to store and distribute container images for deployment; that is why Amazon ECR is preferable here.

Amazon EKS is intended to run Kubernetes-based container workloads using an AWS-managed control plane because eKS integrates managed Kubernetes cluster control plane services with AWS infrastructure; in this design the new outcome is to store and distribute container images for deployment, which calls for Amazon ECR rather than Amazon EKS or the earlier EC2 instance store.

AWS Fargate is intended to run container tasks without managing EC2 container hosts because fargate supplies serverless compute for supported ECS and EKS workloads; in this design the new outcome is to store and distribute container images for deployment, which calls for Amazon ECR rather than AWS Fargate or the earlier EC2 instance store.

Public subnet is intended to place resources in a subnet with a route toward an internet gateway because a public subnet has a route to an attached internet gateway where applicable; in this design the new outcome is to expose and manage HTTP REST or WebSocket APIs, which calls for Amazon API Gateway rather than Public subnet or the earlier Internet gateway.

Amazon VPC is intended to create a logically isolated IP network for AWS resources because a VPC defines virtual networking boundaries and hosts configurable subnets and routing; in this design the new outcome is to expose and manage HTTP REST or WebSocket APIs, which calls for Amazon API Gateway rather than Amazon VPC or the earlier Internet gateway.

Internet gateway primarily supports the earlier need to enable internet routing for resources in a VPC with suitable configuration because an internet gateway connects a VPC to the internet for eligible public addressing paths, but the new task is to expose and manage HTTP REST or WebSocket APIs; that is why Amazon API Gateway is preferable here.

Amazon OpenSearch Service is intended to search and analyze indexed logs and document collections because openSearch Service supports search analytics and log exploration; in this design the new outcome is to analyze image or video content for objects and visual attributes, which calls for Amazon Rekognition rather than Amazon OpenSearch Service or the earlier Amazon Transcribe.

Amazon Transcribe primarily supports the earlier need to convert spoken audio into a text transcript because transcribe provides managed automatic speech recognition, but the new task is to analyze image or video content for objects and visual attributes; that is why Amazon Rekognition is preferable here.

Amazon SageMaker AI is intended to build train and deploy custom machine learning models because sageMaker AI offers managed tooling for ML data preparation training and inference workflows; in this design the new outcome is to analyze image or video content for objects and visual attributes, which calls for Amazon Rekognition rather than Amazon SageMaker AI or the earlier Amazon Transcribe.

AWS CloudWatch dashboards is intended to inspect selected operational measurements in one view because dashboards visualize metrics and alarms for monitored systems; in this design the new outcome is to manage operational tasks across supported compute fleets, which calls for AWS Systems Manager rather than AWS CloudWatch dashboards or the earlier Hybrid cloud deployment.

AWS service APIs is intended to interact programmatically with AWS using supported service requests because AWS APIs expose actions and resource operations through defined programmatic interfaces; in this design the new outcome is to manage operational tasks across supported compute fleets, which calls for AWS Systems Manager rather than AWS service APIs or the earlier Hybrid cloud deployment.

Hybrid cloud deployment primarily supports the earlier need to retain some local systems while extending selected workloads into AWS because hybrid environments combine on-premises resources with cloud services and connectivity, but the new task is to manage operational tasks across supported compute fleets; that is why AWS Systems Manager is preferable here.

RDS Multi-AZ configuration is intended to improve database availability using a standby in another Availability Zone because RDS Multi-AZ options support failover when the primary or supporting infrastructure fails; in this design the new outcome is to enable managed point-in-time recovery for an eligible RDS database, which calls for RDS automated backups rather than RDS Multi-AZ configuration or the earlier RDS read replica.

Database on Amazon EC2 is intended to retain direct control of database host OS and engine installation because self-managing a database on EC2 leaves patching, backups and high availability largely to the customer; in this design the new outcome is to enable managed point-in-time recovery for an eligible RDS database, which calls for RDS automated backups rather than Database on Amazon EC2 or the earlier RDS read replica.

RDS read replica primarily supports the earlier need to offload eligible read queries from a relational database primary because read replicas provide copies usable to serve read traffic with replication behavior, but the new task is to enable managed point-in-time recovery for an eligible RDS database; that is why RDS automated backups is preferable here.

Amazon Route 53 DNS primarily supports the earlier need to direct clients to an endpoint using managed domain name resolution because Route 53 answers DNS queries according to hosted-zone records and routing policies, but the new task is to redirect DNS responses when monitored endpoint health changes; that is why Route 53 failover routing is preferable here.

Availability Zone is intended to separate redundant application components across isolated facilities within one Region because an Availability Zone is one or more discrete data centers with independent infrastructure in a Region; in this design the new outcome is to redirect DNS responses when monitored endpoint health changes, which calls for Route 53 failover routing rather than Availability Zone or the earlier Amazon Route 53 DNS.

AWS Region is intended to select a geographic area in which to place a workload because an AWS Region is a distinct geographic location containing multiple Availability Zones; in this design the new outcome is to redirect DNS responses when monitored endpoint health changes, which calls for Route 53 failover routing rather than AWS Region or the earlier Amazon Route 53 DNS.