EDB PostgreSQL Professional study plan

Install a PostgreSQL cluster from source or packages and configure postgresql.conf, pg_hba.conf, and pg_ident.conf deliberately rather than by default

Map every physical file location (data directory, WAL, tablespaces, pg_wal, pg_xact) and explain the role of each to a colleague in under five minutes

Create roles, databases, schemas, and grants that demonstrate the difference between LOGIN, CREATEDB, CREATEROLE, SUPERUSER, and role inheritance

Set up a repeatable lab script that rebuilds the cluster from scratch so later weeks start from a known state

Execute and interpret VACUUM, ANALYZE, and autovacuum behavior by forcing bloat and then measuring it with pg_stat_user_tables and pg_stat_progress_vacuum

Configure WAL sizing, checkpoint_completion_target, and max_wal_size, then observe checkpoint behavior in the server log

Diagnose and resolve a simulated transaction ID wraparound warning using age(datfrozenxid) queries

Practice routine maintenance: REINDEX, CLUSTER, and pg_dump/pg_restore of a single schema

Read EXPLAIN (ANALYZE, BUFFERS) output line by line and identify the plan node responsible for the largest cost

Create targeted indexes (B-tree, partial, expression, covering with INCLUDE) and prove each one changes a plan for the better

Reproduce a bad row estimate, then fix it with extended statistics or a manual statistics target increase

Tune three server parameters (work_mem, shared_buffers, effective_cache_size) and measure the effect on the same workload

Use pg_stat_statements to rank the top queries by total time and by mean time

Diagnose lock contention by querying pg_locks and pg_stat_activity, then resolve a deliberately introduced deadlock

Identify and terminate long-running or idle-in-transaction sessions safely and explain the consequences of each action

Configure and validate parallel query settings and observe when the planner declines to parallelize

Explain how the planner chooses between nested loop, hash join, and merge join and demonstrate each choice

Configure streaming replication with a primary and a hot standby and verify replication lag under pg_stat_replication

Perform a planned switchover and a simulated unplanned failover, documenting the steps and the data-loss window

Set up continuous archiving with archive_command and test point-in-time recovery to a specific timestamp

Take and verify a base backup with pg_basebackup and confirm the restore actually starts

Compare logical replication with physical streaming replication and build a logical replication setup between two databases

Evaluate quorum-based synchronous replication settings and measure the commit latency trade-off

Configure a backup schedule combining pg_dump logical dumps with physical backups and define retention rules

Write and execute a restore drill from backup into a clean cluster and verify row counts against the source

Harden a cluster: enforce scram-sha-256 authentication, restrict pg_hba.conf entries, set connection limits, and audit role privileges

Implement row-level security and column-level grants on a sample multi-tenant schema and verify isolation with test sessions

Plan and execute a migration from another database engine, including type mapping, sequence handling, and validation of row counts

Configure TLS for client connections and confirm the cipher and certificate in use

Describe the components of the EDB Postgres AI platform and where each fits in an operational lifecycle, keeping vendor claims separate from verified facts

Practice monitoring workflows: query pg_stat_activity, pg_stat_database, and log-based alerting to detect an incident

Review weak areas from earlier weeks and re-run the corresponding labs without notes

Complete two timed mixed-domain practice sets and analyze every wrong answer