The CLF-C02 groups its service knowledge under Cloud Technology and Services, the largest content area of the exam, and analytics is the category candidates most often skip. That is a mistake: AWS’s analytics portfolio shows up in scenario questions (“which service lets you run SQL directly on data in S3 without managing servers?”), and it is one of the easier categories to master because you only need recognition-level familiarity — what each service does and when you would reach for it, not how to configure it.
The trick is to stop memorizing services in isolation and instead learn them as stages of one data pipeline: data is ingested, stored, cataloged and prepared, queried or processed, and finally visualized. Once you can place a service on that pipeline, the exam’s “which service?” questions almost answer themselves. This guide walks the whole pipeline — Amazon Kinesis, Amazon S3, AWS Glue, Amazon Athena, Amazon Redshift, Amazon EMR, and Amazon QuickSight — with the decision cues that map each exam scenario to the right answer.
Where Analytics Fits on the CLF-C02
You will not be asked to write a query or tune a cluster. The exam tests whether you can recognize the purpose of a service from a plain-English description and distinguish confusable pairs (Athena vs Redshift, Glue vs EMR, Kinesis vs a queue). That is the same recognition skill the core services guide builds for compute, storage, and databases — analytics is simply the category most people under-study. Master the pipeline model below and you turn a neglected topic into free points.
The Data Pipeline Mental Model
Almost every AWS analytics architecture follows the same five stages. Hold this table in your head and you have a slot for every service:
| Stage | What happens | Lead AWS service(s) |
|---|---|---|
| Ingest | Get data in — streaming or batch | Amazon Kinesis, Amazon Data Firehose |
| Store | Land raw data cheaply and durably | Amazon S3 (the data lake) |
| Catalog & prepare | Discover, clean, transform, and catalog data | AWS Glue, AWS Lake Formation |
| Query & process | Ask questions of the data | Amazon Athena, Amazon Redshift, Amazon EMR |
| Visualize | Turn results into dashboards for humans | Amazon QuickSight |
The rest of this guide takes the stages in order.
Ingest: Amazon Kinesis
Amazon Kinesis is AWS’s family for working with real-time streaming data — data that arrives continuously in high volume, such as clickstreams, application logs, IoT sensor readings, or financial ticks. Its two most exam-relevant members:
- Amazon Kinesis Data Streams — ingests and stores streaming records so one or more applications can consume and process them in real time.
- Amazon Data Firehose (formerly Kinesis Data Firehose) — the simplest way to load streaming data into destinations like Amazon S3, Amazon Redshift, or Amazon OpenSearch Service, with optional transformation on the way.
The single exam cue for Kinesis is the word real-time (or “streaming,” “continuous”). If a scenario describes processing data as it arrives rather than in scheduled batches, Kinesis is the answer.
Kinesis vs SQS trap: both move data between components, but Kinesis is for real-time analytics on streaming data with multiple consumers and replay; Amazon SQS is a message queue for decoupling application components. A “process this stream of events for analytics” scenario is Kinesis; a “decouple these microservices” scenario is SQS.
Store: Amazon S3 as the Data Lake
Once data is ingested it needs somewhere cheap, durable, and virtually unlimited to live. On AWS that is Amazon S3, which acts as the data lake — a central repository that stores structured, semi-structured, and unstructured data in its raw form. S3’s durability, low cost, and tiered storage make it the natural landing zone that every downstream analytics service reads from. The mechanics of S3 (buckets, storage classes, durability) are covered in the storage services guide; for analytics, the one thing to internalize is that S3 is the data lake other services query and process.
Catalog & Prepare: AWS Glue and Lake Formation
Raw data in S3 is not yet useful — it needs to be discovered, cataloged, and often cleaned or reshaped. That is AWS Glue: a serverless data-integration (ETL — extract, transform, load) service. Glue can crawl your data sources, infer their schema, and populate the AWS Glue Data Catalog (a central metadata store that Athena, Redshift, and EMR all read), then run serverless ETL jobs to transform data with no clusters to manage.
The exam cues for Glue are “serverless ETL,” “data catalog,” “prepare/transform data,” and “no infrastructure to manage.”
AWS Lake Formation sits one layer up: it helps you build, secure, and govern a data lake on top of S3 and Glue, centralizing fine-grained access permissions so you can control who sees which data. Recognition-level cue: “quickly set up and secure a data lake.”
Query: Amazon Athena
Amazon Athena is a serverless, interactive query service that lets you run standard SQL directly against data in Amazon S3 — no servers, no clusters, no loading step. You point Athena at your S3 data (using the Glue Data Catalog for schema), write a query, and pay only for the data each query scans.
Athena is the answer whenever a scenario emphasizes ad-hoc SQL on S3 with no infrastructure: “analysts want to occasionally query log files in S3 using SQL without setting up a database.” Its serverless, pay-per-query model makes it ideal for infrequent or exploratory analysis.
Query at Scale: Amazon Redshift
Amazon Redshift is AWS’s fully managed, petabyte-scale cloud data warehouse, purpose-built for complex analytical queries and business intelligence over large volumes of structured data. Where a transactional database like RDS is optimized for many small reads and writes, Redshift uses columnar storage and massively parallel processing to run heavy aggregations and joins fast. A Redshift Serverless option removes cluster sizing for spiky or unpredictable workloads. Redshift also appears in the database services guide as the “data warehouse” answer — here it is the analytics engine for structured, repeated, high-performance reporting.
Athena vs Redshift — the classic CLF-C02 pair. Both run SQL, so the exam loves to make you choose. Athena = serverless, ad-hoc/occasional queries directly on S3, pay per query, zero setup. Redshift = a provisioned (or serverless) data warehouse for frequent, complex, high-performance analytics over structured data at scale. Occasional queries on raw S3 files → Athena. A dedicated warehouse powering daily BI dashboards → Redshift.
Process Big Data: Amazon EMR
Amazon EMR (Elastic MapReduce) is a managed big-data platform for running open-source frameworks — Apache Spark, Hadoop, Hive, Presto, and more — on clusters, at massive scale. It is the tool for large-scale data processing and transformation, machine-learning data prep, and custom big-data workloads where you want the power and flexibility of those frameworks.
Glue vs EMR trap: both process/transform big data with Spark, but Glue is serverless ETL with no clusters to manage (choose it for straightforward, managed data integration), while EMR gives you managed clusters you configure and control (choose it for large-scale, custom, or framework-heavy processing). “No servers to manage, just transform my data” → Glue; “run a large Spark/Hadoop cluster for heavy custom processing” → EMR.
Visualize: Amazon QuickSight
The last stage turns query results into something a human can act on. Amazon QuickSight is AWS’s serverless business intelligence (BI) service for building interactive dashboards and visualizations. It connects to sources like Athena, Redshift, S3, and RDS, scales to many users, and includes natural-language querying (QuickSight Q) so business users can ask questions in plain English. The exam cue is unmistakable: “dashboards,” “visualizations,” “business intelligence,” “reports for business users” → QuickSight.
A Reference Analytics Pipeline
Putting the whole category together, a typical AWS analytics flow reads left to right across the pipeline:
Streaming data Data lake Catalog/ETL Query/Process Visualize
[ Kinesis / Firehose ] ───────► [ Amazon S3 ] ──► [ AWS Glue + Catalog ] ──► [ Athena / Redshift ] ──► [ QuickSight ]
└► [ Amazon EMR (Spark/Hadoop) ]Streaming events land in S3 via Kinesis/Firehose (batch data is loaded directly); Glue catalogs and transforms it; analysts query it ad-hoc with Athena or through a Redshift warehouse, while EMR handles heavy custom processing; and QuickSight surfaces the results as dashboards. You will not need to design this on the CLF-C02 — but recognizing where a described service sits in this flow is exactly what the exam rewards.
How the CLF-C02 Frames Analytics Questions
| Scenario cue | Right service | Why |
|---|---|---|
| ”Process continuous, real-time streaming data” | Amazon Kinesis | Real-time streaming ingestion/processing |
| ”Central, low-cost repository for all raw data” | Amazon S3 (data lake) | Durable object storage as the lake |
| ”Serverless ETL to prepare and catalog data” | AWS Glue | Serverless data integration + Data Catalog |
| ”Quickly set up and secure a data lake” | AWS Lake Formation | Build/secure/govern the lake |
| ”Run SQL on S3 data with no servers, pay per query” | Amazon Athena | Serverless interactive query |
| ”Data warehouse for complex BI over structured data” | Amazon Redshift | Managed columnar warehouse |
| ”Run Spark/Hadoop for large-scale processing” | Amazon EMR | Managed big-data frameworks |
| ”Interactive dashboards and visualizations for the business” | Amazon QuickSight | Serverless BI/visualization |
| ”Search and analyze logs at scale” | Amazon OpenSearch Service | Search and log analytics |
Common Mistakes to Avoid
| Mistake | The correction |
|---|---|
| Choosing Redshift for occasional, ad-hoc S3 queries | That’s Athena — serverless, pay-per-query, no setup |
| Choosing Athena for a high-performance BI warehouse | That’s Redshift — a managed data warehouse for frequent, complex analytics |
| Confusing AWS Glue with Amazon EMR | Glue = serverless ETL, no clusters; EMR = managed clusters for big-data frameworks |
| Picking SQS for streaming analytics | Streaming/real-time analytics is Kinesis; SQS is a decoupling message queue |
| Thinking QuickSight stores or queries data | QuickSight visualizes; it connects to Athena, Redshift, S3, and others |
| Treating S3 as “just storage” in analytics | On AWS, S3 is the data lake that analytics services read from |
How to Lock This In Before Exam Day
Analytics is a recognition category, and recognition is built by repetition. The most reliable way to make “run SQL on S3 with no servers → Athena” a reflex is to answer the same style of scenario question dozens of times until the service names surface automatically. Sailor.sh’s AWS Cloud Practitioner mock exam bundle is aligned to the CLF-C02 blueprint and drills the Cloud Technology and Services domain — including these analytics discriminations — under timed conditions. To try the format first, work through the free CLF-C02 practice questions and the full Cloud Practitioner practice set.
To round out your service knowledge, pair this with the core services overview, the database services guide (where Redshift’s warehousing role sits), and the storage services guide (the S3 data lake). For the big picture, start with the CLF-C02 exam guide for 2026, see how the marks split in the domains breakdown, and schedule prep with the Cloud Practitioner study plan.
Frequently Asked Questions
What is the difference between Amazon Athena and Amazon Redshift?
Athena is a serverless query service that runs SQL directly on data in Amazon S3 with no infrastructure and pay-per-query pricing — ideal for ad-hoc or occasional analysis. Redshift is a managed data warehouse for frequent, complex, high-performance analytics over large volumes of structured data. Use Athena for quick queries on S3; use Redshift as a dedicated warehouse powering ongoing BI.
Is AWS Glue the same as Amazon EMR?
Both can transform big data with Apache Spark, but Glue is serverless — you run ETL jobs and catalog data with no clusters to manage. EMR provisions managed clusters running frameworks like Spark, Hadoop, and Hive, giving you more control for large-scale or custom processing. Choose Glue for managed, straightforward ETL; choose EMR for heavy, framework-specific big-data workloads.
When would I use Amazon Kinesis instead of a batch process?
Use Amazon Kinesis when data arrives continuously and you need to ingest or analyze it in real time — clickstreams, logs, IoT telemetry, or live metrics. Batch processes handle data in scheduled chunks; Kinesis handles it as it streams in.
What does Amazon QuickSight do?
Amazon QuickSight is AWS’s serverless business intelligence service for building interactive dashboards and visualizations. It connects to sources such as Athena, Redshift, S3, and RDS, scales to many users, and offers natural-language querying through QuickSight Q. It is the “turn data into dashboards for business users” service.
How deep is analytics knowledge tested on the CLF-C02?
Only at recognition level. You need to identify the purpose of each analytics service and distinguish the confusable pairs (Athena vs Redshift, Glue vs EMR, Kinesis vs SQS). You do not need to write queries, design pipelines, or configure clusters.
What is a data lake on AWS?
A data lake is a centralized repository that stores all your data — structured, semi-structured, and unstructured — in its raw form. On AWS, Amazon S3 is the foundation of the data lake, and services like AWS Glue, Athena, Redshift, and EMR read from and process data stored there. AWS Lake Formation helps build and secure it.
Conclusion
AWS analytics stops being intimidating the moment you see it as a single pipeline: Kinesis ingests streaming data, S3 stores it as the data lake, Glue catalogs and transforms it, Athena and Redshift query it (ad-hoc versus warehouse), EMR handles heavy big-data processing, and QuickSight visualizes the results. For the CLF-C02 you need only to recognize each service’s job and win the confusable pairs. Learn the pipeline table and the decision cues above, drill them against realistic questions, and one of the most-skipped exam categories becomes a dependable pocket of easy marks.