XiangpengHao

XiangpengHao / probex

Public

Low-friction Linux profiler with integrated web viewer

20
1
100% credibility
Found Feb 19, 2026 at 15 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Rust
AI Summary

Probex is a straightforward Linux tool that profiles running programs by capturing CPU usage, scheduling, system calls, and IO events, then displays interactive visualizations like timelines, flame graphs, and statistics in a built-in web viewer.

How It Works

1
🔍 Spot a slow program

You notice your Linux app running sluggishly and decide to investigate its performance.

2
📥 Get probex ready

You quickly download or install the profiler with a simple command.

3
🚀 Run and capture

You launch your program through probex, and it effortlessly records everything it does.

4
📈 Viewer launches

A colorful dashboard opens automatically, showing timelines and graphs of activity.

5
🔍 Explore the data

You zoom into processes, check CPU heatmaps, IO patterns, and stack traces.

✅ Fix bottlenecks

You pinpoint the slowdowns and speed up your program with confidence.

Sign up to see the full architecture

4 more

Sign Up Free

Star Growth

See how this repo grew from 15 to 20 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is probex?

Probex is a low-friction Linux profiler in Rust that uses eBPF probes to trace processes out-of-the-box. Run `sudo probex -- your-command` to capture CPU samples via frame pointers, syscalls like read/write/mmap/io_uring, sched switches, page faults, and more into a Parquet file. It auto-launches an integrated web viewer at http://localhost:8080 with flamegraphs, process timelines, IO/memory stats, and histograms.

Why is it gaining traction?

No scripting needed—Nix, Cargo, or binaries just work, unlike perf or bpftrace. Delivers accurate user stacks without DWARF unwinding, plus a Rust-built web viewer rivaling Snapshots or PySpur with interactive viz for latencies, allocs, and process trees. Low-overhead sampling at 1999Hz keeps it practical for prod-like debugging.

Who should use this?

Rust devs profiling async IO in Tokio or io_uring apps, kernel hackers chasing page faults or sched issues, ops engineers dissecting multi-process workloads like databases. Perfect for Linux users wanting quick insights into bottlenecks without raw perf dumps.

Verdict

Solid for ad-hoc Linux profiling despite low maturity (15 stars, 1.0% credibility score)—great README, binaries, and docs make it try-worthy now. Early stage means watch for edge cases; contribute if you need ARM or more events.

(187 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.