HOW POSTGRES
WORKS

How a table really lives on disk: 8KB pages and tuples.

Find one row among millions without reading them all.

Seq scan vs index scan — how Postgres decides.

Keys, foreign keys, CHECK/UNIQUE — how the database refuses bad data.

Store, query, and index semi-structured data with GIN.

Row versions, snapshots, and how transactions stay isolated.

Dead tuples, reclaiming space, autovacuum, and xid wraparound.

Row locks, lock waits, deadlock detection, and how to avoid them.

Write-ahead logging, crash recovery, checkpoints, replication.

shared_buffers, cache hits vs disk reads, and clock-sweep eviction.

How a plan actually runs: a tree of nodes pulling rows one at a time.

Find it, read the plan, fix it — the performance workflow.

Process-per-connection, max_connections, and why you need a pooler.

Splitting a big table into partitions, and partition pruning.

Streaming WAL to standbys, sync vs async, lag, and failover.

Roles, GRANT, and policies that filter rows per tenant — Supabase's core.

Trace a single query through every stage you've learned.