Documentation
The handbook is organised by concern first and by engine second. If you are new here, start with About DBPilot and Choosing a Database.
Introduction
Vocabulary, trade-offs and selection criteria that the rest of the handbook assumes.
Architecture
How database clusters are put together, and what each topology costs you.
Production Best Practices
Engine-independent practices for running a database that other teams depend on.
PostgreSQL
Configuration, indexing, vacuum, replication and recovery for PostgreSQL clusters.
MySQL
InnoDB internals, replication topologies and operational practices for MySQL.
MariaDB
Where MariaDB diverges from MySQL, and how Galera changes the operational model.
Redis and Valkey
Memory behaviour, persistence, clustering and caching failure modes. Applies to Valkey unless stated otherwise.
ClickHouse
MergeTree design, insert patterns, replication and the diagnostics that matter in production.
MongoDB
Document modelling, index design, replica set behaviour and sharding operations.
Cassandra
Ring topology, query-first data modelling, tunable consistency and anti-entropy operations.
ScyllaDB
The shard-per-core execution model and how it changes tuning compared to Cassandra.
Distributed SQL
Consensus-replicated SQL engines: CockroachDB and YugabyteDB in production.
Time Series
Retention, downsampling, compression and cardinality control for metrics workloads.
Elasticsearch and OpenSearch
Index design, cluster sizing and lifecycle management for Lucene-based search clusters.
SQLite and RocksDB
Embedded engines: where they fit, what they cannot do, and how to operate them.
Operations
Backup, recovery, failover and upgrade procedures that apply to every engine.
Performance
Diagnosing latency and throughput problems, and forecasting the next bottleneck.
Monitoring
Metrics, logs, traces and alerts that tell you a database is degrading before users do.
Security
Authentication, authorization, encryption and audit practices for database estates.