developer-tools postgres testing-frameworks go-programming

Pgtestdb's template cloning approach to testing is fast

brandur.org

August 1, 2026

3 min read

🔥🔥🔥🔥🔥

46/100

Summary

pgtestdb is a Go/Postgres testing package that utilizes Postgres template databases for efficient testing. Cloning a template database is significantly faster than migrating a test database from scratch or using Docker-based techniques.

Key Takeaways

pgtestdb utilizes Postgres template databases for fast cloning, significantly reducing setup time compared to traditional migration methods.
The average setup time for pgtestdb cloning is approximately 98.4 ms, while creating and migrating schemas takes around 99.4 ms.
Overall test suite execution with schema-based methods runs about 3.5 times faster than using pgtestdb, due to optimizations in schema reuse.
Reusing schemas can potentially reduce test setup time to 10-20 ms, making it more efficient for large test suites.

Read original article

View story →

Coverage & discussions

Hacker News · brandur.org

Pgtestdb's template cloning approach to testing is fast ↗

Community discussions

Hacker News discussion ↗

Historical count: 5 · 5 saved in sample · Historical sample

Analysis ready

Community Sentiment

Positive

Positives

Using template databases for testing is a game-changer — speeds up the process dramatically, especially when you have a large schema.
The performance gains are noticeable, particularly with a 400-500 table schema; this method really enhances efficiency.
pgtestdb's approach to providing a clean database quickly is impressive, especially when migrations run only once, saving time across multiple tests.
Running PostgreSQL on a ramdisk can take the speedup even further, making the cloning process lightning fast and ideal for intensive testing.

Concerns

Some commenters are skeptical about the effort involved, suggesting that using fakes for tests might be easier and cheaper, especially for lower layers of testing.
Concerns were raised about the potential fragility of fakes in complex database operations, which can lead to misleading test results.
There’s a worry that the 'dirty db' testing approach could lead to ambiguous failures, making debugging more complicated.