CheckOps home

Tools directory

P

CheckOps

Parca

Parca is an Apache-2.0 continuous profiling project that analyses CPU and memory usage down to the line number over time, using an eBPF agent.

Visit website

Latest public signals

Not refreshed yet. Refresh reads the public site and, when the entry has a GitHub repository, the latest release.

Visit Website

Introduction

Parca provides continuous profiling for analysis of CPU and memory usage, down to the line number and over time, to save infrastructure cost, improve performance and increase reliability. The Parca Agent uses eBPF to profile all processes on a node without code changes.

The source code is at parca-dev/parca on GitHub under the Apache-2.0 licence, with about 5,000 stars; the most recent commit was on 11 November 2026.

Features

  1. eBPF agent: Profiles without instrumentation.
  2. Line-level detail: Pinpoint hot code.
  3. Time-series profiles: Compare over time.
  4. pprof compatible: Standard profile format.
  5. Label-based queries: Prometheus-style labels.

Pros

  1. Free to use: Open source under Apache-2.0.
  2. Maintained: Commits as recent as 11 November 2026.
  3. Community: About 5,000 GitHub stars.

Cons

  1. Linux only agent: eBPF needs Linux.
  2. Storage and retention: Plan for data growth.

Deploy the Parca server and Parca Agent DaemonSet using the manifests in the docs, then open the Parca UI to explore flame graphs.

Resources:

Pricing

Parca is free and open source under the Apache-2.0 licence.

  1. Performance engineers reducing CPU cost.
  2. Kubernetes platforms adding profiling.
  3. Not a fit for metrics or logs.

Conclusion

Parca brings eBPF-based continuous profiling with line-level detail.

Alternatives

Other tools on CheckOps that cover similar ground. Open a page, or compare one with Parca side by side.