PostgreSQL Field Notes

Notes for the problems that show up after launch: bad plans, awkward migrations, index debt, vacuum pressure, replica lag, and the small decisions that make PostgreSQL easier to operate.

PostgreSQL Topic Hubs

PostgreSQL Guide (Pillar)

pgvector Monitoring Query Optimizer Sizing Slow Query Monitoring Query Monitoring

PostgreSQL Tools

HNSW Tuning Lab Plan Autopsy Index Rollout

Tool Comparisons

Compare PostgreSQL monitoring tools

Browse by topic

pgvector and RAG Indexes Slow Queries Locks and Transactions Backups and Durability Replication and WAL Connections and Pooling CPU, Memory, and I/O Vacuum and Bloat Cloud Providers

January 31, 20269 min read

WAL Monitoring in Postgres: What to Watch Before Disk Becomes the Story

WAL problems usually look like disk problems too late. Monitor generation rate, checkpoints, archiving, replication lag, and slot retention before pg_wal owns the incident.

January 30, 20268 min read

NUMERIC vs DOUBLE PRECISION in Postgres: The Difference That Hides in Money

Use NUMERIC when exactness is the product contract. Use DOUBLE PRECISION when measurement error is already part of the domain and speed matters more than decimal identity.

January 29, 20267 min read

Full-Text Search in Postgres: Good Enough, Until It Isn't

Postgres full-text search is great for the cases it covers and frustrating for the cases it doesn't. Knowing where the line is saves a year of trying to make it do what Elasticsearch does.

January 28, 20268 min read

CITEXT in Postgres: Case-Insensitive Text Without the LOWER Trap

CITEXT is a good fit when a column is almost always compared case-insensitively. It is not a universal Unicode answer, and it is not free.

January 27, 20266 min read

Range Types in Postgres: The Built-in Solution to Overlap Problems

Range types model time windows, numeric ranges, and any "from-to" data. They support overlap queries natively and combine with EXCLUDE constraints to prevent invalid data.

January 26, 20269 min read

Materialized Views in Postgres: Cheap Reads, Expensive Refreshes

Materialized views are excellent when stale-enough reads are worth a scheduled rebuild. They are painful when teams expect incremental freshness or forget the refresh lock model.

January 25, 20266 min read

Rollup Tables in Postgres: The Pattern That Replaces Most BI Queries

Rollup tables are pre-aggregated summaries of source data, kept up to date via triggers or scheduled jobs. They turn dashboard queries from seconds into milliseconds.

January 24, 20268 min read

Postgres Checkpoint Tuning: The Latency Spike Nobody Sees Coming

Checkpoint problems show up as periodic latency spikes. The work is smoothing writes, sizing WAL, and proving checkpoints are the thing users feel.

January 23, 20267 min read

Time-Series Data in Postgres: Honest Tradeoffs Before You Reach for TimescaleDB

Postgres handles time-series data better than people assume. The decision to add a specialized extension should come from observed limits, not preemptive optimization.

January 22, 20269 min read

Incremental Aggregation in Postgres: When Materialized Views Are Not Enough

Postgres materialized views refresh a result set; they do not give you automatic incremental aggregation. For high-churn dashboards, you need deltas, watermarks, and late-data rules.

January 21, 20265 min read

Postgres High CPU: Where the Cycles Actually Go

When Postgres CPU is high, the question is not whether the server is fast enough. It is which queries are using the cycles and whether they should be.

January 20, 20266 min read

Postgres High I/O: Diagnosing Where Disk Goes Wrong

High I/O is not a storage verdict. It is a workload question: which query, table, index, vacuum, checkpoint, or spill is reading and writing the bytes?