mirror of
https://github.com/hashicorp/nomad-device-nvidia
synced 2026-04-05 19:02:05 +00:00
No description
- Go 97.2%
- HCL 1.5%
- Makefile 0.7%
- Shell 0.6%
4e0ced426dBumps [github.com/hashicorp/nomad](https://github.com/hashicorp/nomad) from 1.11.2 to 1.11.3. - [Release notes](https://github.com/hashicorp/nomad/releases) - [Changelog](https://github.com/hashicorp/nomad/blob/main/CHANGELOG.md) - [Commits](https://github.com/hashicorp/nomad/compare/v1.11.2...v1.11.3) --- updated-dependencies: - dependency-name: github.com/hashicorp/nomad dependency-version: 1.11.3 dependency-type: direct:production update-type: version-update:semver-patch ... Signed-off-by: dependabot[bot] <support@github.com> Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com> |
||
|---|---|---|
| .github | ||
| .release | ||
| cmd | ||
| nvml | ||
| scripts | ||
| version | ||
| .copywrite.hcl | ||
| .gitignore | ||
| .go-version | ||
| CHANGELOG.md | ||
| device.go | ||
| device_test.go | ||
| fingerprint.go | ||
| fingerprint_test.go | ||
| go.mod | ||
| go.sum | ||
| LICENSE | ||
| Makefile | ||
| README.md | ||
| stats.go | ||
| stats_test.go | ||
Nomad Nvidia Device Plugin
This repository provides an implementation of a Nomad
device plugin
for Nvidia GPUs.
Behavior
The Nvidia device plugin uses
NVML bindings to get data
regarding available Nvidia devices and will expose them via
Fingerprint
RPC. GPUs can be excluded from fingerprinting by setting the ignored_gpu_ids
field (see below). Plugin sends statistics for fingerprinted devices periodically.
The plugin detects whether the GPU has Multi-Instance GPU (MIG) enabled.
When enabled all instances will be fingerprinted as individual GPUs that can be addressed accordingly.
Config
The plugin is configured in the Nomad client's
plugin block:
plugin "nvidia" {
config {
ignored_gpu_ids = ["uuid1", "uuid2"]
fingerprint_period = "5s"
}
}
The valid configuration options are:
ignored_gpu_ids(list(string):[]): list of GPU UUIDs strings that should not be exposed to nomadfingerprint_period(string:"1m"): interval to repeat the fingerprint process to identify possible changes.