https://github.com/ordo-one/swift-parca
Low-overhead continuous profiling for Swift on Server production environments
https://github.com/ordo-one/swift-parca
docker linux parca performance performance-analysis swift swift-server
Last synced: 8 months ago
JSON representation
Low-overhead continuous profiling for Swift on Server production environments
- Host: GitHub
- URL: https://github.com/ordo-one/swift-parca
- Owner: ordo-one
- License: apache-2.0
- Created: 2025-08-26T08:19:52.000Z (10 months ago)
- Default Branch: main
- Last Pushed: 2025-08-26T14:00:15.000Z (10 months ago)
- Last Synced: 2025-10-21T07:52:30.433Z (8 months ago)
- Topics: docker, linux, parca, performance, performance-analysis, swift, swift-server
- Language: Swift
- Homepage:
- Size: 32.2 KB
- Stars: 51
- Watchers: 5
- Forks: 1
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# swift-parca
`swift-parca` supports continuous low-overhead profiling of Swift Server applications at scale, for always-on tracking of performance issues in production.
`swift-parca` is a fully packaged solution that extends [Parca](https://www.parca.dev), an open-source continuous profiling tool that embeds [opentelemetry-ebpf-profiler](https://github.com/parca-dev/opentelemetry-ebpf-profiler) for creating low overhead samples, with native support for Swift symbol demangling.
Parca consists of an agent that runs on each node that is under monitoring as well as a server that aggregates all the profiling data for analysis using familiar tools such as flamegraphs and sample overviews.
## Sample output
### Example of reviewing `swift-build` stacks
### Example of `swift-nio` benchmark
## 🚀 Getting Started
`swift-parca` is provided as a single, pre-configured Docker container image that bundles Parca and all necessary dependencies, to simplify profiling of Swift applications at scale.
To begin profiling your Swift application, follow these simple steps:
1. Start the Parca Server
Use the provided `docker-compose.yml` to bring up the Parca server:
```bash
docker compose up -d
```
This starts the Parca server, listening on HTTP port 7070 for UI and profiling data from agents.
2. Start the Parca Agent
On each server running your Swift workloads, deploy the `parca-agent`:
```bash
export REMOTE_STORE_ADDRESS=:7070
docker-compose -f docker-compose.agent.yml up -d
```
## Documentation
For further setup and usage instructions, please refer to the [Parca Documentation](https://www.parca.dev/docs/overview).
## Performance Overhead
[Parca agent](https://github.com/parca-dev/parca-agent) samples [19 times per second](https://www.parca.dev/docs/parca-agent-design/#obtaining-raw-data) per core, only the stack addresses are aggregated in eBPF maps; the rest of the processing is sending them and binaries to the server, where demangling/symbolicatition happens on demand when a trace is viewed. In practice this results in neglible overhead on the profiled application, that can be sub 1%.
## Background
Parca normally uses a [Go demangler](https://github.com/ianlancetaylor/demangle) which [lacks Swift symbol](https://github.com/ianlancetaylor/demangle/issues/16) demangling.
This project addresses this limitation by configuring Parca to delegate address-to-line lookups and symbol demangling to a custom `swift-addr2line` embedded tool. This tool leverages the native Swift runtime function for demangling, ensuring accurate and human-readable stack traces, while relying on a Rust-based [`gimli` addr2line](https://github.com/gimli-rs/addr2line) (or to [`llvm-addr2line`](https://llvm.org/docs/CommandGuide/llvm-addr2line.html)) for actual address lookup.
This direct access of the native Swift runtime function is needed until [SE-262](https://github.com/swiftlang/swift-evolution/blob/main/proposals/0262-demangle.md) is formally integrated into the standard library.
## Using llvm-addr2line
If you want to use [`llvm-addr2line`](https://llvm.org/docs/CommandGuide/llvm-addr2line.html) instead of the `gimli-rs/addr2line`, uncomment the line containing `apt-get install -y llvm-19` in the Dockerfile.
This installs the LLVM tools, making the binary available at `/usr/bin/llvm-addr2line-19`.
Then you need to update the `docker-compose.yml` environment:
```bash
environment:
- ADDR2LINE_PATH=/usr/bin/llvm-addr2line-19
```
## Requirements
The container is packaged for Linux x86_64 and ARM64. It is possible to run entirely under a Linux VM (with kernel 5.4+) under Apple sillicon.