This question tests understanding of no-code data integration in AWS. Amazon AppFlow is built to move data between SaaS apps and AWS services without writing custom code. It offers prebuilt connectors, lets you define flows that extract data from a source, optionally transform it, and load it into a destination. You can schedule flows or trigger them automatically, and AWS handles the plumbing—authentication, incremental transfers, retries, and error handling—so no ETL scripting is required. The other services don’t focus on no-code data transfers: a queue service like SQS is for decoupling components, an event router like EventBridge handles event-driven patterns, and Managed Blockchain is for building blockchain networks. Therefore, the best choice is Amazon AppFlow.

Automatically adjusting the number of EC2 instances in response to demand is the purpose of Amazon EC2 Auto Scaling. It lets you define an Auto Scaling group with a desired capacity, minimum, and maximum size, and it continuously monitors workload metrics (like CPU utilization or request counts). When load rises, it launches new instances to handle the extra traffic; when load falls, it terminates instances to avoid wasted capacity. It also replaces unhealthy instances and distributes traffic through a load balancer, keeping the application available and responsive. This is the best fit for ensuring you have enough EC2 capacity because it directly manages the fleet size, scales automatically, and works within defined safety bounds. The other options don’t provide this automatic capacity management for EC2: launching and running individual instances without scaling, offering cost analytics rather than scaling, or representing a different service model that abstracts away EC2 capacity (not specifically about scaling EC2 instances).

Amazon DocumentDB (with MongoDB compatibility) is a fully managed service designed to run MongoDB workloads in the cloud. It provides a MongoDB-compatible API, so applications can use the same drivers and tools they’re already using, while AWS handles the heavy lifting of operations. You set up a cluster, and AWS takes care of provisioning, patching, backups, and monitoring. Storage scales automatically, and you can scale compute by adjusting instance types and add read replicas to handle higher read throughput. It integrates with AWS security and monitoring features (VPC, IAM, CloudWatch), making it easy to deploy, operate, and grow MongoDB-compatible databases without managing the underlying infrastructure. DynamoDB, Neptune, and ElastiCache serve different purposes (general NoSQL, graph databases, and in-memory caching, respectively) and aren’t MongoDB-compatible managed options, which is why DocumentDB is the fit for MongoDB-compatible needs.

On-demand, high-quality text translation is provided by Amazon Translate. It’s a fully managed neural machine translation service designed to translate text between many languages quickly and accurately, with real-time and batch options that you can call directly from your applications via API. This makes it ideal for adding multilingual capabilities without building or maintaining your own translation infrastructure. The other options don’t fit because the AWS CLI is just a command-line tool for managing AWS resources, not a translation service; SageMaker is a platform for building and deploying ML models (you could train a translation model there, but it isn’t a ready-made on-demand translation service); and Textract focuses on extracting text from documents rather than translating it.

Centralized log management in AWS is handled by a service built specifically for gathering and working with log data. CloudWatch Logs lets you collect logs from EC2 instances (via the CloudWatch Logs agent or the CloudWatch Agent) and from other sources, store them in central log groups, and access them for viewing, searching, and analysis. You can retain logs for as long as you need, export them to S3 for archival, and run queries or use Logs Insights to extract meaningful information. It also integrates with CloudWatch Metrics and Alarms so you can alert on patterns found in logs. AWS CloudTrail logs API activity rather than general application or OS logs, and AWS License Manager is for managing software licenses, not logging.