Database Administration

About this course

This mid-level course takes you from writing SQL to running databases that other people depend on, the core of the database administrator role. The first module covers design: normalising tables to third normal form, choosing keys, data types and constraints for the workload, and deciding when a relational database or a document, key-value, wide-column or graph store fits better. The second module is performance: indexing strategy and composite indexes, reading query plans, rewriting slow queries, and how transactions, isolation levels, locks and deadlocks behave under concurrency. The third module is operations on a real PostgreSQL server that you run yourself in Docker or as a native install: configuration files, roles, logical backups and tested restores, point-in-time recovery, and replication with safe failover. The last module covers security and monitoring: least-privilege roles with GRANT and REVOKE, watching connections, slow queries and bloat with pg_stat_statements and the statistics views, and planning zero-downtime schema migrations with expand, backfill and contract. Design and performance lessons have SQL exercises that run in your browser on SQLite; operations lessons are hands-on steps on a practice server with practice data only. Most ideas transfer to MySQL, SQL Server and managed cloud databases, and lessons point out where engines differ. Employers usually also look for a degree or equivalent experience, vendor certifications and on-call experience, which you build on the job.