Load Testing Made Easy with Gatling

on YouTube ↗35m

" summary

Builds a small Scala 3 GraphQL service backed by Postgres and exercises it with Gatling simulations that ramp 0 → 20 users over 30s, hold constant 30s, then ramp down, generating random valid queries with ScalaCheck generators and sending them via Gatling. Initial test on ~500 companies and a few thousand employees fails with ~95% connection timeouts until missing foreign-key indexes are added; after indexing requests finish under 500ms. A second simulation fetching 5–10 companies shows ~10× lower throughput and several responses over 2–4s.

AI-written summary · about

" chapters (10)

1. 0:00Load testing with Gatling
2. 1:12Test scenario and API setup
3. 10:49Scala service for testing
4. 12:13Gatling simulation phases
5. 15:13Dynamic GraphQL queries with ScalaCheck
6. 23:39First load test and results
7. 25:26Fixing DB indexes for performance
8. 27:17Second GraphQL query test
9. 31:35Open vs closed workload models
10. 33:28Summary

" related talks

• 34mPerformance tests with GatlingAndrzej Ludwikowski · code.talks 2017
• 46mContinuous Performance: Load Testing with GatlingTim van Eijndhoven · J-Fall 2015
• 50mPerformance testing with GatlingAndrzej Michałowski · Confitura 2015
• 12mWeb load testing with GatlingJonathan Araña · DrupalCamp Spain 2017
• 21mGATLING DISTILLEDAndrzej Ludwikowski · Scalar Conference 2017
• 38mPerformance Testing sucks! How to get started testing RESTful APIs and GraphQLLars Wolff · code.talks 2017
• 36mAutomated Performance Testing of PostgresGreg Burek · PostgresOpen 2014
• 21mCrunching Millions of Requests for Breakfast: A Tale of Performance Testing ProwessSergey Kurnikov · Heapcon 2024