x64arm64 at 2 vCPU4 vCPU8 vCPU
docker pullsecondslower is betterarm64 · 8 vCPU
RunsOn tmpfs pulls an image fastest.
RunsOn tmpfs to Ubicloud: 3.9× apart.
- RunsOntmpfs m9g.2xlarge · 32 GB 4.5s
- RunsOn median of 3 machines 4.9s
- Avrea 32 GB 5.3s
- Warpbuild 32 GB 8.0s
- Blacksmith 32 GB 11.4s
- Namespace median of 2 machines 13.1s
- Ubicloud standard · 32 GB 17.6s
- WarpbuildApple silicon 12 vCPU · xfast (Apple M4 Pro) · 28 GB 7.6s not ranked
- AWS CodeBuild no arm64 runner
- boat no arm64 runner
- GitHub no 8 vCPU arm64 runner
- NamespaceApple silicon no 8 vCPU arm64 runner
The same image pulled on every runner, registry to disk: the download, then the layers unpacked, so the disk and the CPU count too.