Categorygithub.com/hashicorp/nomaddevicesgpunvidia
package
0.12.12
Repository: https://github.com/hashicorp/nomad.git
Documentation: pkg.go.dev
Versions
1
Dependencies
13
Dependents
25
Files
715 SLOC

# README

This package provides an implementation of nvidia device plugin

Behavior

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. Plugin sends statistics for fingerprinted devices every stats_period period.

Config

The configuration should be passed via an HCL file that begins with a top level config stanza:

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 nomad
  • fingerprint_period (string: "1m"): interval to repeat the fingerprint process to identify possible changes.

# Packages

# Functions

NewNvidiaDevice returns a new nvidia device plugin.

# Constants

Mebibytes.
number of errors.
number of errors.
number of errors.
Attribute names and units for reporting Fingerprint output.
Mebibytes.
Nvidia-container-runtime environment variable names.
Attribute names for reporting stats output.
Celsius degrees.

# Variables

PluginConfig is the nvidia factory function registered in the plugin catalog.
PluginID is the nvidia plugin metadata registered in the plugin catalog.

# Structs

Config contains configuration information for the plugin.
NvidiaDevice contains all plugin specific data.