Tracing Java's Hidden Performance Traps
summary
Covers practical techniques for locating Java backend performance bottlenecks using distributed tracing, metrics, and method profiling. Explains how to interpret flame graphs and examines common causes such as connection pool starvation, expensive Aspects, slow network calls, CPU hotspots, lock contention, and Virtual Thread pinning.
AI-written summary · about
related talks
46mProfiling your Java Application A Beginner’s Guide
44mJVM Profiling under the Hood
49mDistributed Tracing in Java? OpenTelemetry for the Rescue!
32mWhat's the Performance Overhead? Answering the Biggest Question in Tracing
45mMastering JVM Profiling: Advanced Techniques and Real-Life Case Studies
46mHow to Properly Blame Things for Causing Latency
46mHitchhiker's Guide to Performance Reloaded for Java 15
48mFantastic Performance and Where to Find It