Parca
parca.dev
- Free plan
- Linux
- Web
- API
- Self-hosted
What Parca is
Parca is a free, open-source profiling system that continuously collects and stores CPU, memory, I/O, and other profiles for later querying. It is self-hosted, and Parca Agent uses eBPF to sample user-space and kernel-space stack traces. Profiles can be sent over gRPC or gathered from targets over HTTP, and Parca accepts valid pprof-formatted profiles. Its query tools include label selectors, a profiling-specific engine, and comparisons across dimensions such as processes or code versions. The web interface can visualize results as Icicle Graphs, and a Grafana plugin can display profiles stored in Parca. Automatic profiling supports languages including C, C++, Go, Rust, .NET, Java, Node.js, PHP, Ruby, and Python, along with other listed languages. The project describes sampling as suitable for continuous production use with very little overhead; its FAQ reports observed Agent overhead below 1% CPU and below 200 Mb of memory, depending on active targets. Parca says it receives profiling statistics and function-name metadata, not binaries or source code. The components are available under the Apache 2 License. Agent use requires Linux kernel 4.18 or newer and privileged access to load eBPF programs.
Who it suits
Parca suits developers and operations teams that want to profile CPU, memory, and other application behavior over time while keeping deployment self-hosted. It is especially relevant to teams able to provide the required Linux kernel version and privileged Agent access.
Good to know, and what to watch
- Continuously collects CPU, memory, I/O, and other profiles
- Supports gRPC push and HTTP pull collection
- Query profiles by labels and compare dimensions
- Web UI visualizes profiles with Icicle Graphs
- Open-source components under Apache 2 License
- Agent requires Linux kernel 4.18 or newer
- Agent needs privileged access to load eBPF programs
- Parca focuses on sampling profiling
Plans
Every plan and priceParca Open Source
Open source
Getting started
- Open the Parca website
- Choose a documented deployment route: binaries, containers, Kubernetes, or Snaps
- Install Parca Agent on Linux kernel 4.18 or newer
- Grant the Agent kernel access and privileged permissions for eBPF
- Push profiles through gRPC or pull them from targets over HTTP
- Query profiles in the web interface or visualize them with Grafana
Limits to check
Parca Agent requires Linux kernel 4.18 or newer and privileged access to load eBPF programs. Reported overhead varies with the number of actively profiled targets.
The verdict
Parca offers self-hosted profile collection and querying, with a web UI and Grafana visualization option. Teams should confirm that the Linux and privileged-access requirements fit their deployment environment.
Questions people ask
Is Parca free?
Yes. The Parca Open Source plan is free, and the project is open source.
Which platforms does Parca support?
Its listed platforms are API, Linux, self-hosted, and web.
What languages can Parca Agent profile automatically?
Supported languages include C, C++, Go, Rust, .NET, Deno, Erlang, Java, Julia, Node.js, Wasmtime, PHP, Ruby, and Python.
How can profiles enter Parca?
Profiles can be pushed through gRPC or pulled from targets through HTTP. Parca also supports valid pprof-formatted profiles.
Can Parca work with Grafana?
Yes. A Grafana plugin can visualize profiles stored in Parca.
What the maker publishes
- Agent
- Parca Agent is an eBPF-based whole-system profiler. source
- Analysis
- Parca supports label-selector queries, a profiling-specific query engine, and comparisons across label dimensions such as processes or code versions. source
- Commercial API
- The project identifies Polar Signals as the only commercial provider of Parca-compatible APIs, through Polar Signals Cloud. source
- Data collection
- Profiles can be pushed to Parca through gRPC or pulled from targets through HTTP. source
- Data handling
- Parca says it receives profiling statistics and function-name metadata, not users’ binaries or source code. source
- Deployment
- The project documents installation using binaries, containers, Kubernetes, and Snaps. source
- Features
- Features include a multi-dimensional data model, label-selector queries, a profiling-specific query engine, built-in storage, and support for pushing and pulling profiles. source
- Grafana
- Profiles stored in Parca can be visualized using the Grafana plugin. source
- Integration
- A Grafana plugin can visualize profiles stored in Parca. source
- Languages
- The agent’s automatic profiling supports languages including C, C++, Rust, Go, Ruby, Node.js, Python, Java, .NET, Perl, and PHP. source
- Linux requirement
- Parca Agent requires Linux kernel version 4.18 or newer. source
- Linux requirement
- The quick start says the Parca Agent binary works only on Linux. source
- Maker
- Polar Signals says it was founded in 2020. source
- Open source
- Parca’s components are available under the Apache 2 License. source
- Overhead
- The FAQ reports observed Agent overhead of less than 1% CPU and less than 200 Mb memory, depending on the number of actively profiled targets. source
- Product
- Parca continuously collects, stores, and makes CPU, memory, I/O, and other profiles available for querying over time. source
- Profiler
- Parca Agent uses eBPF to capture user-space and kernel-space stack traces with low overhead. source
- Profiling
- Parca focuses on sampling profiling, which the project says can run continuously in production with very little overhead. source
- Security
- The agent needs kernel access and privileged-user permissions to load eBPF programs. source
- Security
- The project says the Agent source is open source, it is written in Go, and its binaries and container images are fully reproducible; the Agent needs privileged access to load eBPF programs. source
- Support
- Polar Signals offers paid support for users of Parca’s open-source project. source
- Supported formats
- Parca supports any valid pprof-formatted profile. source
- Visualization
- Parca’s web UI can query profiles and visualize them using Icicle Graphs, described as upside-down Flame Graphs. source
Checked against the maker's pages on 7 Oct 2026.