Runup of AWS Services Databases & Data Platform

From your first managed database to lakehouses, streaming and migration - 35 AWS data services, one problem at a time.
What you'll learn

---

Pick the right AWS data store on purpose: relational, key-value, document, graph, time-series, warehouse or lake.
Know when to load data and when to query it where it already sits - Redshift vs Athena vs Spectrum vs lakehouse.
Build a streaming path, and explain the hot-key mistake that breaks Redshift, DynamoDB, Kinesis and Kafka alike.
Plan a live database or mainframe migration that needs no outage, using DMS, MGN, DataSync and AWS Transform.
Requirements

---

Basic SQL familiarity helps. Part 1 (Core Infrastructure) is the recommended starting point but is not required.
Description

---

Disclaimer: This course contains the use of artificial intelligence.
Runup of AWS Services series covers the 35 services AWS gives you for storing, querying, moving and making sense of data.
Data is where AWS gets confusing fastest. There are at least nine ways to store a table, four ways to query files you already have, and three separate services that all claim to move a database - and the documentation explains what each one does without ever telling you which one you need. This course fixes that by teaching every service through the problem it was invented to solve, so choosing between them stops being a memory exercise.
It starts with the database itself. Module 1 builds a four-rung ladder - your own database on an instance, then RDS, then Aurora, then Aurora DSQL - and attaches one rule to it: climb only as far as your actual problem pushes you. Most companies stop at the second rung and are right to, and the course says so out loud.
From there the question changes shape. Module 2 asks whether your data should be a table at all, and answers it seven ways: DynamoDB to fetch by key at any size, ElastiCache for the same answer repeatedly, MemoryDB for when you cannot afford to lose it, DocumentDB for nested documents, Neptune for connections several steps out, Keyspaces for existing Cassandra code, Timestream for measurements over time. You leave able to run down that list, not able to recite seven feature sets.
Module 3 is the largest and the strongest: one escalating question - how much machinery does an answer deserve? Redshift, Athena, Glue, Lake Formation, EMR, OpenSearch, S3 Tables and S3 Vectors each take the next step in a single argument. The course then turns outward to the people who actually read the data, with Quick Suite, DataZone, Data Exchange, Clean Rooms and Entity Resolution; starts the clock with Kinesis, Data Firehose, MSK and Managed Flink; and ends on the hardest part of any real project, which is moving what already exists without the outage nobody will authorise.
Along the way it names the mistakes that repeat. The hot partition key is the same error whether it is called a distribution key in Redshift, a partition key in DynamoDB, a shard key in Kinesis or a partition key in Kafka - and the lessons point back at each other so you recognise one pattern instead of meeting four unrelated services. Hot-and-cold tiering gets the same treatment across log retention, search indices, storage classes and vectors.
Everything is taught in plain language with heavy illustration and minimum jargon, across 48 animated lessons and roughly three and a half hours of video. A quiz follows every service, and a 30-question final assessment closes the course by making you choose between services - the thing a per-service quiz structurally cannot ask.
Who this course is for

---

Engineers and analysts who can write SQL but freeze when asked which AWS data service to put it in front of.
Anyone preparing for the AWS Data Engineer Associate exam who wants the reasoning rather than a list of service limits.
Developers who have outgrown one database and need to know what the next choice will actually cost them.
Teams planning a move off an existing database, warehouse or mainframe who need to compare the available routes.
Homepage

---