Early Performance Testing: Finding News You Can Use
Talkat CAST 2014 · on Tuesday, 2014-08-12· at 15:30· 10th fl Rosenthal Pavillion, Kimmel Center for University Life·
summary
Argues for fast, cheap, repeatable checks instead of one costly end-of-project simulation: run small synthetic workloads (notably a 10x10 test or 1x100 runs, even a 17-second slam) daily to detect regressions, use Fiddler or profilers to find root causes, keep a single-server calibration to track deltas, and run soak tests to catch leaks. Recommends JMeter or simple home-grown tools, CI integration (Jenkins/TeamCity), playback of production logs, and targeted tests for login, session creation, message bus capacity and database hotspots.
AI-written summary · about
chapters (16)
- 0:00Start
- 0:48Iterative Development
- 1:55The Risks That We Address with Performance Testing
- 2:28Scalability
- 3:41Test for Capacity
- 10:40Indictment of the Traditional Simulation Test
- 12:06Make Automation Easier To Live with
- 13:32Horizontal Scalability
- 13:55Calibration
- 18:33Soap Tests
- 31:27Logging Response Times for Common Activities
- 31:49Reporting on Performance Results
- 37:02The Simulation Test
- 38:07Business Logic Layer
- 38:27Authentication Store
- 47:00Jmeter for Starting with Performance Testing
related talks
40mAutomated Performance Testing In Preproduction
29mProgressive Testing to Meet the Performance Imperative
33mTESTING IN PRODUCTION
57mOpen Forum on Performance Testing
41mContinuous Performance Testing in DevOps
41mMartijn Goossens: Become a Performance Test Master - the Non-Technical Side of Things
57mAdvanced Testing Techniques: Tips from the trenches
01h 02mThe Art & Joy of Testing in Production