Java ProfilerLow-overhead Java profiling for Kubernetes incidents

A focused profiler for HotSpot services on Kubernetes: opt-in collection, async-profiler/JFR-derived evidence, ClickHouse storage, and a UI built for Java incident diagnosis.

Quickstart

Analyze a Service

GitHub

Opt-in by default

Profiling is opt-in through Kubernetes metadata, collected node-locally, and retained for 7 days or less.

Real Java evidence

CPU, Wall Clock, Java I/O wait, GC, allocation summaries, lock delay, thread, deadlock, status, and ingestion evidence stay tied to one service and time range.

Own the profiling stack

No required Pyroscope, Parca, or Grafana backend. async-profiler data lands in ClickHouse and a self-owned UI.

Fast time to first signal

The first usable path is status, then CPU or allocation summary, then Wall Clock, I/O, GC, locks, or ingestion when the incident needs deeper evidence.

Real allocation profile analysis from the acceptance environment

3-minute path

Open Quickstart and enable profiling with Kubernetes metadata.
Open the target service, then check status first to confirm the JVM was accepted.
Move from cpu or memory to wall, io, gc, locks, and ingestion to get from symptom to evidence.

For service owners

Quickstart: enable profiling and read your first service profile.
Performance Analysis Manual: read CPU, Wall Clock, Java I/O wait, GC, allocation summary, lock, deadlock, target status, profile evidence guidance, and ingestion evidence.
Java Profiling Runbook: enable temporary or continuous profiling for a Kubernetes workload.

For platform operators

Deployment and Operations: install, secure, operate, upgrade, and troubleshoot the profiler.
Real Profiling Acceptance: prove CPU, Wall Clock, Java I/O wait, GC, allocation, lock, ClickHouse, UI, and ingestion behavior before shipping changes.

For contributors

Contributing: run local checks, build docs, and execute real acceptance.
Architecture: understand the collector, backend, ClickHouse store, contracts, and web UI.
E2E Automation Guide: run browser and real Kubernetes acceptance flows.
Profiling Contracts: inspect stable payload and configuration contracts.
GitHub Issues: report bugs and request changes.

Trust and localization

Use the English / 简体中文 switch in the top bar when you want the localized path.
The real screenshots on this page come from the acceptance environment, not mocked UI state. The allocation screenshot shows the current Allocation Summary and flamegraph workflow.

Local preview

Run these commands from the repository root:

bash

cd docs
npm install
npm run docs:dev

Build the publishable static site:

bash

cd docs
npm run docs:build

Java ProfilerLow-overhead Java profiling for Kubernetes incidents

Opt-in by default

Real Java evidence

Own the profiling stack

Fast time to first signal

3-minute path ​

For service owners ​

For platform operators ​

For contributors ​