[go: up one dir, main page]
logo
search
Back to Pixie
COLLAPSE ALL
  • Home
  • About Pixie
    • Pixie Overview
    • Data Sources
    • How Pixie uses eBPF
    • Roadmap
    • FAQ
    • Troubleshooting
  • Installing Pixie
    • Requirements
  • Using Pixie
    • Using the Live UI
    • Using the CLI
    • Using the API
  • Tutorials
    • Pixie 101
      • Network Monitoring
      • Infra Health
      • Service Performance
      • Database Query Profiling
      • Request Tracing
      • Continuous Application Profiling
      • Kafka Monitoring
  • Reference
    • Architecture
WebsiteSlack Us

Continuous Application Profiling

This tutorial will demonstrate how to use Pixie's Always-On Profiling feature to investigate a spike in CPU utilization, using a flamegraph to identify a performance issue within the application code.

Pixie's continuous profiler currently supports Go, C++, Rust and Java. For best results, run Java applications with -XX:+PreserveFramePointer. Other language support coming soon.

Prerequisites

You will need a Kubernetes cluster with Pixie installed. If you do not have a cluster, you can create a minikube cluster and install Pixie using one of our install guides.

Java Support

Support for Java profiling is included by default in Pixie.

For best results, run Java applications with -XX:+PreserveFramePointer.

Setup