Merge pull request #65 from Monokaix/dev

migrate vgpu to external project

.github/workflows/main.yml

@@ -34,5 +34,3 @@ jobs:
  - run: make ubuntu20.04
  - run: TAG_VERSION="${BRANCH_NAME}" make push-tag
  - run: make push-latest
- - run: make vgpu
- - run: TAG_VERSION="${BRANCH_NAME}" make push-vgpu-tag

Makefile

@@ -1,4 +1,4 @@
-# Copyright (c) 2023, NVIDIA CORPORATION. All rights reserved.
+# Copyright (c) 2020, NVIDIA CORPORATION. All rights reserved.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.
@@ -20,8 +20,7 @@
 
 DOCKER ?= docker
 REGISTRY ?= volcanosh
-VERSION ?= latest
-TAG_VERSION ?= 1.0.0
+VERSION ?= 1.0.0
 
 ##### Public rules #####
 
@@ -39,24 +38,11 @@ push-latest:
  $(DOCKER) tag "$(REGISTRY)/volcano-device-plugin:$(VERSION)-ubuntu20.04" "$(REGISTRY)/volcano-device-plugin:latest"
  $(DOCKER) push "$(REGISTRY)/volcano-device-plugin:latest"
 
-push-tag:
- $(DOCKER) tag "$(REGISTRY)/volcano-device-plugin:$(VERSION)-ubuntu20.04" "$(REGISTRY)/volcano-device-plugin:$(TAG_VERSION)"
- $(DOCKER) push "$(REGISTRY)/volcano-device-plugin:$(TAG_VERSION)"
-
-push-vgpu-tag:
- $(DOCKER) tag "$(REGISTRY)/volcano-vgpu-device-plugin:$(VERSION)-ubuntu20.04" "$(REGISTRY)/volcano-vgpu-device-plugin:$(TAG_VERSION)"
- $(DOCKER) push "$(REGISTRY)/volcano-vgpu-device-plugin:$(TAG_VERSION)" 
-
 ubuntu20.04:
- $(DOCKER) build --pull \
+ $(DOCKER) build --network=host --pull \
  --tag $(REGISTRY)/volcano-device-plugin:$(VERSION)-ubuntu20.04 \
  --file docker/amd64/Dockerfile.ubuntu20.04 .
 
-vgpu:
- $(DOCKER) build --pull \
- --tag $(REGISTRY)/volcano-vgpu-device-plugin:$(VERSION)-ubuntu20.04 \
- --file docker/amd64/Dockerfile.vgpu-ubuntu20.04 .
-
 centos7:
  $(DOCKER) build --pull \
  --tag $(REGISTRY)/volcano-device-plugin:$(VERSION)-centos7 \

README.md

@@ -73,41 +73,13 @@ We will be editing the docker daemon config file which is usually present at `/e
 Once you have enabled this option on *all* the GPU nodes you wish to use,
 you can then enable GPU support in your cluster by deploying the following Daemonset:
 
-VGPU:
-```
-$ kubectl create -f volcano-vgpu-device-plugin.yml
-```
-
-GPU-SHARE (**Will be deprecated in volcano v1.9**):
 ```shell
 $ kubectl create -f volcano-device-plugin.yml
 ```
 
 **Note** that volcano device plugin can be configured. For example, it can specify gpu strategy by adding in the yaml file ''args: ["--gpu-strategy=number"]'' under ''image: volcanosh/volcano-device-plugin''. More configuration can be found at [volcano device plugin configuration](https://github.com/volcano-sh/devices/blob/master/doc/config.md).
 
-### Running VGPU Jobs
-
-VGPU can be requested by both set "volcano.sh/vgpu-number" and "volcano.sh/vgpu-memory" in resource.limit
-
-```shell script
-$ cat <<EOF | kubectl apply -f -
-apiVersion: v1
-kind: Pod
-metadata:
- name: gpu-pod1
-spec:
- containers:
- - name: cuda-container
- image: nvidia/cuda:9.0-devel
- command: ["sleep"]
- args: ["100000"]
- resources:
- limits:
- volcano.sh/vgpu-number: 2 # requesting 1 gpu cards
- volcano.sh/vgpu-memory: 3000
-EOF
-```
-### Running GPU Sharing Jobs (**Will be deprecated in volcano v1.9**)
+### Running GPU Sharing Jobs (Without memory isolation)
 
 NVIDIA GPUs can now be shared via container level resource requirements using the resource name volcano.sh/gpu-memory:
 
@@ -144,7 +116,7 @@ spec:
 > **WARNING:** *if you don't request GPUs when using the device plugin with NVIDIA images all
 > the GPUs on the machine will be exposed inside your container.*
 
-### Running GPU Number Jobs (**Will be deprecated in volcano v1.9**)
+### Running GPU Number Jobs (Without number isolation)
 
 NVIDIA GPUs can now be requested via container level resource requirements using the resource name volcano.sh/gpu-number:
 
@@ -170,7 +142,7 @@ EOF
 
 Please note that:
 - the device plugin feature is beta as of Kubernetes v1.11.
-- the gpu-share device plugin is alpha and is missing the following features, and will be deprecated in volcano v1.9
+- the Volcano device plugin is alpha and is missing
  - More comprehensive GPU health checking features
  - GPU cleanup features
  - GPU hard isolation
@@ -180,16 +152,19 @@ The next sections are focused on building the device plugin and running it.
 
 ### With Docker
 
-#### Deploy as DaemonSet:
+#### Build
+```shell
+$ make ubuntu20.04.
+```
 
-GPU-SHARE:
+#### Run locally
 ```shell
-$ kubectl create -f nvidia-device-plugin.yml
+$ docker run --security-opt=no-new-privileges --cap-drop=ALL --network=none -it -v /var/lib/kubelet/device-plugins:/var/lib/kubelet/device-plugins nvidia/k8s-device-plugin:{version}
 ```
 
-VGPU:
+#### Deploy as DaemonSet:
 ```shell
-$ kubectl create -f nvidia-vgpu-device-plugin.yml
+$ kubectl create -f nvidia-device-plugin.yml
 ```
 
 # Issues and Contributing

docker/amd64/Dockerfile.ubuntu20.04

@@ -40,4 +40,4 @@ ENV NVIDIA_DRIVER_CAPABILITIES=utility
 
 COPY --from=build /go/src/volcano.sh/devices/volcano-device-plugin /usr/bin/volcano-device-plugin
 
-ENTRYPOINT ["volcano-device-plugin"]
+ENTRYPOINT ["volcano-device-plugin"]