4.3AI/ML and Analytics Services (Overview)

Domain 3 also expects you to recognize the main categories of AWS services beyond compute, storage, networking, and databases. Two that come up often nowadays are AI/machine learning and analytics. The goal here is not implementation detail but being able to connect "for this purpose, which AWS service?"

4.3.1AI and machine-learning services

When you want to build, train, and deploy models yourself, the core is Amazon SageMaker. Separately, a family of prebuilt AI services lets you add advanced capabilities by calling an API—no ML expertise required.

• Amazon Rekognition: image/video analysis (objects, faces, moderation).
• Amazon Comprehend: extract sentiment, key phrases, and entities from text (NLP).
• Amazon Lex: build conversational bots (chat/voice).
• Amazon Polly (text-to-speech) and Amazon Transcribe (speech-to-text).
• Amazon Translate (translation), Amazon Textract (extract text/tables from documents), Amazon Kendra (intelligent enterprise search).

Recently generative AI services joined the main categories (recognizing name and purpose is enough). Amazon Bedrock provides managed access to foundation models (generative AI) from multiple providers, and Amazon Q is a generative AI assistant for business and development tasks.

4.3.2Analytics services

• Amazon Athena: query data in S3 directly with SQL (serverless).
• Amazon Redshift: a data warehouse for large-scale analytics.
• Amazon EMR: big-data processing with Spark/Hadoop.
• Amazon Kinesis: ingest and process real-time streaming data.
• AWS Glue: serverless ETL (extract, transform, load) and data catalog.
• Amazon QuickSight: BI dashboards (visualization). Amazon OpenSearch Service: search and log analytics.

4.3.3In-scope services in other categories (overview)

Because CLF-C02 tests "name and purpose" broadly, also know these categories at an awareness level. Networking (more): Amazon API Gateway to publish/manage APIs, AWS Global Accelerator to speed up global traffic, AWS PrivateLink for private connectivity between VPCs and services, AWS Transit Gateway to interconnect many VPCs/on-prem, and AWS Client VPN to connect individual users to a VPC. Application integration: Amazon EventBridge for event routing, Amazon Simple Notification Service (SNS) for notifications, Amazon Simple Queue Service (SQS) for queues, and AWS Step Functions for workflows. Business/frontend/IoT: Amazon Simple Email Service (SES) for email, AWS Amplify for frontend development, AWS AppSync for GraphQL APIs, and AWS IoT Core for IoT device connectivity. End user computing: Amazon WorkSpaces virtual desktops, Amazon WorkSpaces Secure Browser for secure browsing, and Amazon AppStream 2.0 for application streaming. Database (more): Amazon DocumentDB (MongoDB-compatible). ML/dev: Amazon SageMaker AI to build ML, AWS CLI for operations, AWS CodeBuild for builds, AWS CodePipeline for deployment automation, and AWS X-Ray for distributed tracing.

4.3.4In-scope migration and transfer services (overview)

Services for migrating from on-premises to AWS are also in scope. AWS Migration Hub is the control tower to plan and track the whole migration; AWS Application Discovery Service discovers and inventories on-prem assets beforehand; Migration Evaluator builds a data-driven cost estimate (business case); AWS Application Migration Service actually moves servers to AWS; AWS Database Migration Service migrates databases (with AWS Schema Conversion Tool for heterogeneous schema conversion); and AWS Snow Family ships large data sets on physical devices when the network is impractical. Map by role: "plan = Migration Hub / discover = Application Discovery Service / estimate = Migration Evaluator / move servers = Application Migration Service / migrate DBs = Database Migration Service + Schema Conversion Tool / bulk physical transfer = Snow Family."

4.3.5Section summary

• AI/ML: build with SageMaker, prebuilt Rekognition/Comprehend/Lex/Polly/Transcribe/Translate/Textract/Kendra, gen AI Bedrock/Amazon Q
• Analytics: Athena (SQL on S3) / Redshift (DWH) / EMR (Spark) / Kinesis (streaming) / Glue (ETL) / QuickSight (BI) / OpenSearch (search)
• Overview: app integration = EventBridge/SNS/SQS/Step Functions, frontend/IoT = Amplify/AppSync/IoT Core/SES, EUC = WorkSpaces/AppStream 2.0, migration = Migration Hub/MGN/DMS/Snow Family
• CLF-C02 is awareness level—matching name to purpose is enough