Lee Barnes: Maximizing the Value of Performance Testing Throughout the Software Development Cycle
summary
Many performance efforts produce misleading results and even harm decisions; real value is the actionable findings derived from measurable objectives and solid planning. The talk shows how to convert vague questions into 95th percentile response time objectives, contrasts protocol vs browser scripting (protocol scales to thousands of users but can take weeks to build; browser scripts are faster to author but need heavy infra), names cloud load generators like OctoPerf BlazeMeter and k6, cites ~3% adoption of unit/microbenchmarks, and urges continuous performance testing throughout the pipeline.
AI-written summary · about
chapters (18)
- 0:00Introduction
- 1:09Performance testing terms
- 3:13Check-box testing pitfalls
- 4:59Pyramid of testing practices
- 5:40Measurable objectives
- 7:48Test execution planning
- 10:16Test environment setup
- 11:44Reliable test scripts
- 13:08Protocol vs browser scripting
- 15:17Script accuracy & data integrity
- 17:06Load generation options
- 18:52Real-time monitoring
- 19:35Analyzing test data
- 23:07Reporting results
- 27:18Testing in pipeline
- 29:27Continuous testing
- 33:04Key takeaways
- 34:49Q&A
related talks
50mCreating Performance Impact in a Changing Tech Landscape
29mProgressive Testing to Meet the Performance Imperative
57mOpen Forum on Performance Testing
12mPerformance In Context
08m"Why caring about good performance?"
49mPerformance Testing
40mPerformance Testing Done Right
01h 00mPerformance In Modern Times: An Evolutionary Journey