Skip to content

Commit

Permalink
Update fake-toolkit-ready and disable all fractional gpu config by de…
Browse files Browse the repository at this point in the history 
…fault (#853)

* Update fake-toolkit-ready.yaml

* Disable fractional gpus config by default

* Update README.md with readme-generator-for-helm

Signed-off-by: chiragjn <[email protected]>

* Update fake-toolkit-ready.yaml

---------

Signed-off-by: chiragjn <[email protected]>
Co-authored-by: chiragjn <[email protected]>
  • Loading branch information
@chiragjn
chiragjn and chiragjn authored Dec 5, 2024
1 parent e8c477d commit 6f9ad1d
Show file tree
Hide file tree
Showing 4 changed files with 44 additions and 16 deletions.
2 changes: 1 addition & 1 deletion charts/tfy-gpu-operator/Chart.yaml
View file
Original file line number Diff line number Diff line change
@@ -1,6 +1,6 @@
apiVersion: v2
name: tfy-gpu-operator
version: 0.1.23
version: 0.1.24
description: "Truefoundry GPU Operator"
maintainers:
- name: truefoundry
Expand Down
6 changes: 4 additions & 2 deletions charts/tfy-gpu-operator/README.md
View file
Original file line number Diff line number Diff line change
Expand Up @@ -55,6 +55,7 @@ Tfy-gpu-operator is a Helm chart that facilitates the deployment and management
| `aws-eks-gpu-operator.dcgmExporter.resources.limits.cpu` | CPU limit for the DCGM Exporter. | `100m` |
| `aws-eks-gpu-operator.dcgmExporter.resources.limits.memory` | Memory limit for the DCGM Exporter. | `1000Mi` |
| `aws-eks-gpu-operator.dcgmExporter.args` | Arguments for the DCGM Exporter. | `["-c","5000"]` |
| `aws-eks-gpu-operator.mig.strategy` | migStrategy for mig node, single or mixed | `none` |

### gcp-gke-standard-driver Configuration for the GKE Standard Nvidia Drivers. This section will only be used when clusterType.gcpGkeStandard is set to true.

Expand Down Expand Up @@ -123,7 +124,7 @@ Tfy-gpu-operator is a Helm chart that facilitates the deployment and management
| `azure-aks-gpu-operator.driver.enabled` | Enable/Disable driver installation. | `false` |
| `azure-aks-gpu-operator.toolkit.enabled` | Enable/Disable nvidia container toolkit installation. | `true` |
| `azure-aks-gpu-operator.toolkit.version` | Version of the toolkit. Note for Aure Linux change `-ubuntu20.04` to `-ubi8`. However at the time of writing Azure Linux only supports V100 and T4 GPUs | `v1.17.2-ubuntu20.04` |
| `azure-aks-gpu-operator.mig.strategy` | migStrategy for mig node, single or mixed | `mixed` |
| `azure-aks-gpu-operator.mig.strategy` | migStrategy for mig node, single or mixed | `none` |
| `azure-aks-gpu-operator.devicePlugin.enabled` | Enable/Disable nvidia device plugin installation. | `true` |
| `azure-aks-gpu-operator.dcgm.enabled` | Enabled/Disable standalone DCGM. | `false` |
| `azure-aks-gpu-operator.dcgm.version` | Image tag for DCGM container. Find all image tags at https://catalog.ngc.nvidia.com/orgs/nvidia/teams/cloud-native/containers/dcgm/tags | `3.3.8-1-ubuntu22.04` |
Expand Down Expand Up @@ -180,6 +181,7 @@ Tfy-gpu-operator is a Helm chart that facilitates the deployment and management
| `civo-talos-gpu-operator.dcgmExporter.resources.limits.cpu` | CPU limit for the DCGM Exporter. | `100m` |
| `civo-talos-gpu-operator.dcgmExporter.resources.limits.memory` | Memory limit for the DCGM Exporter. | `1000Mi` |
| `civo-talos-gpu-operator.dcgmExporter.args` | Arguments for the DCGM Exporter. | `["-c","5000"]` |
| `civo-talos-gpu-operator.mig.strategy` | migStrategy for mig node, single or mixed | `none` |

### generic-gpu-operator Configuration for the GPU Operator. This section will only be used when clusterType.generic is set to true.

Expand Down Expand Up @@ -209,7 +211,6 @@ Tfy-gpu-operator is a Helm chart that facilitates the deployment and management
| `generic-gpu-operator.driver.enabled` | Enable/Disable driver installation. | `true` |
| `generic-gpu-operator.toolkit.enabled` | Enable/Disable nvidia container toolkit installation. | `true` |
| `generic-gpu-operator.toolkit.version` | Version of the toolkit. | `v1.17.2-ubuntu20.04` |
| `generic-gpu-operator.mig.strategy` | migStrategy for mig node, single or mixed | `mixed` |
| `generic-gpu-operator.devicePlugin.enabled` | Enable/Disable nvidia device plugin installation. | `true` |
| `generic-gpu-operator.dcgm.enabled` | Enabled/Disable standalone DCGM. | `false` |
| `generic-gpu-operator.dcgm.version` | Image tag for DCGM container. Find all image tags at https://catalog.ngc.nvidia.com/orgs/nvidia/teams/cloud-native/containers/dcgm/tags | `3.3.8-1-ubuntu22.04` |
Expand All @@ -225,3 +226,4 @@ Tfy-gpu-operator is a Helm chart that facilitates the deployment and management
| `generic-gpu-operator.dcgmExporter.resources.limits.cpu` | CPU limit for the DCGM Exporter. | `100m` |
| `generic-gpu-operator.dcgmExporter.resources.limits.memory` | Memory limit for the DCGM Exporter. | `1000Mi` |
| `generic-gpu-operator.dcgmExporter.args` | Arguments for the DCGM Exporter. | `["-c","5000"]` |
| `generic-gpu-operator.mig.strategy` | migStrategy for mig node, single or mixed | `none` |
3 changes: 2 additions & 1 deletion charts/tfy-gpu-operator/templates/fake-toolkit-ready.yaml
View file
Original file line number Diff line number Diff line change
Expand Up @@ -28,6 +28,7 @@ spec:
- |-
set -ex;
touch /run/nvidia/validations/host-driver-ready;
touch /run/nvidia/validations/driver-ready;
touch /run/nvidia/validations/toolkit-ready;
sleep infinity;
resources:
Expand Down Expand Up @@ -131,4 +132,4 @@ spec:
maxUnavailable: 1
maxSurge: 0
revisionHistoryLimit: 1
{{- end }}
{{- end }}
49 changes: 37 additions & 12 deletions charts/tfy-gpu-operator/values.yaml
View file
Original file line number Diff line number Diff line change
Expand Up @@ -93,8 +93,8 @@ aws-eks-gpu-operator:
- name: nvidia.com/gpu
replicas: 10
name: time-slicing-config
create: true
default: all
create: false
default: ''

## Node Feature Discovery configuration.
node-feature-discovery:
Expand Down Expand Up @@ -336,6 +336,11 @@ aws-eks-gpu-operator:
## @param aws-eks-gpu-operator.dcgmExporter.args Arguments for the DCGM Exporter.
args: ["-c", "5000"]

## MIG Configuration
mig:
## @param aws-eks-gpu-operator.mig.strategy migStrategy for mig node, single or mixed
strategy: none

## MIG Manager configuration.
migManager:
## @skip aws-eks-gpu-operator.migManager.enabled
Expand Down Expand Up @@ -709,7 +714,7 @@ azure-aks-gpu-operator:
## MIG Configuration
mig:
## @param azure-aks-gpu-operator.mig.strategy migStrategy for mig node, single or mixed
strategy: mixed
strategy: none

## Device Plugin configuration.
devicePlugin:
Expand All @@ -736,8 +741,8 @@ azure-aks-gpu-operator:
- name: nvidia.com/gpu
replicas: 10
name: time-slicing-config
create: true
default: all
create: false
default: ''

## DCGM configuration
dcgm:
Expand Down Expand Up @@ -963,6 +968,21 @@ civo-talos-gpu-operator:
value: volume-mounts
- name: DEVICE_ID_STRATEGY
value: index
## @skip civo-talos-gpu-operator.devicePlugin.config
config:
data:
all: ''
time-sliced-10: |-
version: v1
sharing:
timeSlicing:
renameByDefault: true
resources:
- name: nvidia.com/gpu
replicas: 10
name: time-slicing-config
create: false
default: ''

## DCGM configuration
dcgm:
Expand Down Expand Up @@ -1029,6 +1049,11 @@ civo-talos-gpu-operator:
## @param civo-talos-gpu-operator.dcgmExporter.args Arguments for the DCGM Exporter.
args: ["-c", "5000"]

## MIG Configuration
mig:
## @param civo-talos-gpu-operator.mig.strategy migStrategy for mig node, single or mixed
strategy: none

## MIG Manager configuration.
migManager:
## @skip civo-talos-gpu-operator.migManager.enabled
Expand Down Expand Up @@ -1193,11 +1218,6 @@ generic-gpu-operator:
- name: ACCEPT_NVIDIA_VISIBLE_DEVICES_AS_VOLUME_MOUNTS
value: 'true'

## MIG Configuration
mig:
## @param generic-gpu-operator.mig.strategy migStrategy for mig node, single or mixed
strategy: mixed

## Device Plugin configuration.
devicePlugin:
## @param generic-gpu-operator.devicePlugin.enabled Enable/Disable nvidia device plugin installation.
Expand All @@ -1223,8 +1243,8 @@ generic-gpu-operator:
- name: nvidia.com/gpu
replicas: 10
name: time-slicing-config
create: true
default: all
create: false
default: ''

## DCGM configuration
dcgm:
Expand Down Expand Up @@ -1291,6 +1311,11 @@ generic-gpu-operator:
## @param generic-gpu-operator.dcgmExporter.args Arguments for the DCGM Exporter.
args: ["-c", "5000"]

## MIG Configuration
mig:
## @param generic-gpu-operator.mig.strategy migStrategy for mig node, single or mixed
strategy: none

## MIG Manager configuration.
migManager:
## @skip generic-gpu-operator.migManager.enabled
Expand Down

0 comments on commit 6f9ad1d

Please sign in to comment.