Skip to content

rasterandstate/fjord-bench-github

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

3 Commits
 
 
 
 
 
 
 
 

Repository files navigation

fjord-bench-github

The GitHub Actions baseline for rasterstate/fjord-actions.

The Fjord bundle benchmarks its cache/artifact/setup actions on a self-hosted Forgejo runner. The reference point is the same workload built through the GitHub Actions stack with the upstream actions (actions/setup-node, actions/setup-go, dtolnay/rust-toolchain). That has to run on GitHub Actions, so it lives here.

This runs on a self-hosted GitHub runner (same class of hardware as the Fjord runner), so the number isolates the action ecosystem rather than the cloud hardware: a same-hardware comparison of GitHub Actions vs Fjord Actions.

What it does

.github/workflows/bench.yml builds three pinned workloads on a self-hosted Linux runner and records a cold and a warm build time each:

Workload Repo Pinned ref
Rust BurntSushi/ripgrep 14.1.1
Node axios/axios v1.7.9
Go cli/cli v2.62.0

"Warm" is a second build with the toolchain's on-disk caches left intact, i.e. the state actions/cache restores at the start of a steady-state run. That is the fair comparison to the Fjord warm (cache-restored) number.

Results land in results/github.csv in the schema the Fjord harness consumes:

workload,scenario,phase,seconds,runner_os,runner_arch,run_id,timestamp

scenario is always github so the Fjord update-readme.sh can drop the file straight into its results dir.

How the numbers travel

The Fjord bench workflow fetches this CSV over the raw URL each run:

https://raw.githubusercontent.com/rasterandstate/fjord-bench-github/main/results/github.csv

The workloads and refs are kept in lockstep with bench-rust.sh / bench-node.sh / bench-go.sh in the Fjord repo. Change a workload in one place, change it in both.

About

No description, website, or topics provided.

Resources

Stars

0 stars

Watchers

0 watching

Forks

Releases

No releases published

Packages

 
 
 

Contributors

Languages