by Rakesh Menon (@sublimecoder) on Sunday, August 7, 2016
- Full talk
- Technical level
Make people comfortable with tools and methodologies used to diagnose and fix performance issues, memory leaks and high CPU utilization in runtime in production runtime.
We’ll mostly be covering the science behind debugging production performance issues and how not to get overwhelmed by it. To stop trying random solutions to make your application respond well under high load.
- Memory profiling tools (“heapdump” module and Chrome debugger + other alternatives)
- CPU profiling: visualizations like flame graphs
- Log analysis tools and structuring log data to help with the analysis
- System tweaks (tried and tested)