Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

add pod yaml files for power example #129

Merged
merged 1 commit into from
Nov 26, 2023
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
29 changes: 29 additions & 0 deletions examples/power/conf/deploy-cpu/cifar10.yaml
Original file line number Diff line number Diff line change
@@ -0,0 +1,29 @@
apiVersion: v1
kind: Pod
metadata:
name: deckard
spec:
containers:
- name: deckard
image: ghcr.io/simplymathematics/deckard:main
imagePullPolicy: Always
workingDir: /deckard/examples/power/
args: ["python", "-m", "deckard", "--config_file", "cifar10.yaml"]
env:
- name: DECKARD_DEVICE
value: "cpu"
# - name: REDIS_HOST
# value: "redis"
# - name: REDIS_PORT
# value: "6379"
# - name: REDIS_DB
# value: "0"
# - name: REDIS_PASSWORD
# value: ""
volumeMounts:
- mountPath: /deckard/
name: mypvc
volumes:
- name: mypvc
persistentVolumeClaim:
claimName: podpvc
29 changes: 29 additions & 0 deletions examples/power/conf/deploy-cpu/cifar100.yaml
Original file line number Diff line number Diff line change
@@ -0,0 +1,29 @@
apiVersion: v1
kind: Pod
metadata:
name: deckard
spec:
containers:
- name: deckard
image: ghcr.io/simplymathematics/deckard:main
imagePullPolicy: Always
workingDir: /deckard/examples/power/
args: ["python", "-m", "deckard", "--config_file", "cifa100.yaml"]
env:
- name: DECKARD_DEVICE
value: "cpu"
# - name: REDIS_HOST
# value: "redis"
# - name: REDIS_PORT
# value: "6379"
# - name: REDIS_DB
# value: "0"
# - name: REDIS_PASSWORD
# value: ""
volumeMounts:
- mountPath: /deckard/
name: mypvc
volumes:
- name: mypvc
persistentVolumeClaim:
claimName: podpvc
14 changes: 14 additions & 0 deletions examples/power/conf/deploy-cpu/default.yaml
Original file line number Diff line number Diff line change
@@ -0,0 +1,14 @@
num_nodes: 1
cluster_name: k8s-cluster
gpu_type: nvidia-tesla-v100
gpu_count: 1
gpu_driver_version: default
machine_type: n1-standard-2
min_nodes: 1
max_nodes: 1
storage_config: conf/deploy/sclass.yaml
persistent_volume_claim: conf/deploy/pvc.yaml
pod : conf/deploy/pod.yaml
image_project: ubuntu-os-cloud
image_family: ubuntu-2204-lts
mount_directory: /mnt/filestore
29 changes: 29 additions & 0 deletions examples/power/conf/deploy-cpu/mnist.yaml
Original file line number Diff line number Diff line change
@@ -0,0 +1,29 @@
apiVersion: v1
kind: Pod
metadata:
name: deckard
spec:
containers:
- name: deckard
image: ghcr.io/simplymathematics/deckard:main
imagePullPolicy: Always
workingDir: /deckard/examples/power/
args: ["python", "-m", "deckard", "--config_file", "mnist.yaml"]
env:
- name: DECKARD_DEVICE
value: "cpu"
# - name: REDIS_HOST
# value: "redis"
# - name: REDIS_PORT
# value: "6379"
# - name: REDIS_DB
# value: "0"
# - name: REDIS_PASSWORD
# value: ""
volumeMounts:
- mountPath: /deckard/
name: mypvc
volumes:
- name: mypvc
persistentVolumeClaim:
claimName: podpvc
11 changes: 11 additions & 0 deletions examples/power/conf/deploy-cpu/pvc.yaml
Original file line number Diff line number Diff line change
@@ -0,0 +1,11 @@
kind: PersistentVolumeClaim
apiVersion: v1
metadata:
name: podpvc
spec:
accessModes:
- ReadWriteMany
storageClassName: filestore-sc
resources:
requests:
storage: 256Gi
10 changes: 10 additions & 0 deletions examples/power/conf/deploy-cpu/sclass.yaml
Original file line number Diff line number Diff line change
@@ -0,0 +1,10 @@
apiVersion: storage.k8s.io/v1
kind: StorageClass
metadata:
name: filestore-sc
provisioner: filestore.csi.storage.gke.io
volumeBindingMode: Immediate
allowVolumeExpansion: true
parameters:
tier: standard
network: default
32 changes: 32 additions & 0 deletions examples/power/conf/deploy-nvidia-tesla-a100/cifar10.yaml
Original file line number Diff line number Diff line change
@@ -0,0 +1,32 @@
apiVersion: v1
kind: Pod
metadata:
name: deckard
spec:
containers:
- name: deckard
image: ghcr.io/simplymathematics/deckard:main
imagePullPolicy: Always
workingDir: /deckard/examples/power/
args: ["python", "-m", "deckard", "--config_file", "cifar10.yaml"]
env:
- name: DECKARD_DEVICE
nvidia-tesla-a100
# - name: REDIS_HOST
# value: "redis"
# - name: REDIS_PORT
# value: "6379"
# - name: REDIS_DB
# value: "0"
# - name: REDIS_PASSWORD
# value: ""
resources:
limits:
nvidia.com/gpu: 1
volumeMounts:
- mountPath: /deckard/
name: mypvc
volumes:
- name: mypvc
persistentVolumeClaim:
claimName: podpvc
32 changes: 32 additions & 0 deletions examples/power/conf/deploy-nvidia-tesla-a100/cifar100.yaml
Original file line number Diff line number Diff line change
@@ -0,0 +1,32 @@
apiVersion: v1
kind: Pod
metadata:
name: deckard
spec:
containers:
- name: deckard
image: ghcr.io/simplymathematics/deckard:main
imagePullPolicy: Always
workingDir: /deckard/examples/power/
args: ["python", "-m", "deckard", "--config_file", "cifa100.yaml"]
env:
- name: DECKARD_DEVICE
nvidia-tesla-a100
# - name: REDIS_HOST
# value: "redis"
# - name: REDIS_PORT
# value: "6379"
# - name: REDIS_DB
# value: "0"
# - name: REDIS_PASSWORD
# value: ""
resources:
limits:
nvidia.com/gpu: 1
volumeMounts:
- mountPath: /deckard/
name: mypvc
volumes:
- name: mypvc
persistentVolumeClaim:
claimName: podpvc
14 changes: 14 additions & 0 deletions examples/power/conf/deploy-nvidia-tesla-a100/default.yaml
Original file line number Diff line number Diff line change
@@ -0,0 +1,14 @@
num_nodes: 1
cluster_name: k8s-cluster
gpu_type: nvidia-tesla-v100
gpu_count: 1
gpu_driver_version: default
machine_type: n1-standard-2
min_nodes: 1
max_nodes: 1
storage_config: conf/deploy/sclass.yaml
persistent_volume_claim: conf/deploy/pvc.yaml
pod : conf/deploy/pod.yaml
image_project: ubuntu-os-cloud
image_family: ubuntu-2204-lts
mount_directory: /mnt/filestore
32 changes: 32 additions & 0 deletions examples/power/conf/deploy-nvidia-tesla-a100/mnist.yaml
Original file line number Diff line number Diff line change
@@ -0,0 +1,32 @@
apiVersion: v1
kind: Pod
metadata:
name: deckard
spec:
containers:
- name: deckard
image: ghcr.io/simplymathematics/deckard:main
imagePullPolicy: Always
workingDir: /deckard/examples/power/
args: ["python", "-m", "deckard", "--config_file", "mnist.yaml"]
env:
- name: DECKARD_DEVICE
nvidia-tesla-a100
# - name: REDIS_HOST
# value: "redis"
# - name: REDIS_PORT
# value: "6379"
# - name: REDIS_DB
# value: "0"
# - name: REDIS_PASSWORD
# value: ""
resources:
limits:
nvidia.com/gpu: 1
volumeMounts:
- mountPath: /deckard/
name: mypvc
volumes:
- name: mypvc
persistentVolumeClaim:
claimName: podpvc
11 changes: 11 additions & 0 deletions examples/power/conf/deploy-nvidia-tesla-a100/pvc.yaml
Original file line number Diff line number Diff line change
@@ -0,0 +1,11 @@
kind: PersistentVolumeClaim
apiVersion: v1
metadata:
name: podpvc
spec:
accessModes:
- ReadWriteMany
storageClassName: filestore-sc
resources:
requests:
storage: 256Gi
10 changes: 10 additions & 0 deletions examples/power/conf/deploy-nvidia-tesla-a100/sclass.yaml
Original file line number Diff line number Diff line change
@@ -0,0 +1,10 @@
apiVersion: storage.k8s.io/v1
kind: StorageClass
metadata:
name: filestore-sc
provisioner: filestore.csi.storage.gke.io
volumeBindingMode: Immediate
allowVolumeExpansion: true
parameters:
tier: standard
network: default
32 changes: 32 additions & 0 deletions examples/power/conf/deploy-nvidia-tesla-p100/cifar10.yaml
Original file line number Diff line number Diff line change
@@ -0,0 +1,32 @@
apiVersion: v1
kind: Pod
metadata:
name: deckard
spec:
containers:
- name: deckard
image: ghcr.io/simplymathematics/deckard:main
imagePullPolicy: Always
workingDir: /deckard/examples/power/
args: ["python", "-m", "deckard", "--config_file", "cifar10.yaml"]
env:
- name: DECKARD_DEVICE
nvidia-tesla-p100
# - name: REDIS_HOST
# value: "redis"
# - name: REDIS_PORT
# value: "6379"
# - name: REDIS_DB
# value: "0"
# - name: REDIS_PASSWORD
# value: ""
resources:
limits:
nvidia.com/gpu: 1
volumeMounts:
- mountPath: /deckard/
name: mypvc
volumes:
- name: mypvc
persistentVolumeClaim:
claimName: podpvc
32 changes: 32 additions & 0 deletions examples/power/conf/deploy-nvidia-tesla-p100/cifar100.yaml
Original file line number Diff line number Diff line change
@@ -0,0 +1,32 @@
apiVersion: v1
kind: Pod
metadata:
name: deckard
spec:
containers:
- name: deckard
image: ghcr.io/simplymathematics/deckard:main
imagePullPolicy: Always
workingDir: /deckard/examples/power/
args: ["python", "-m", "deckard", "--config_file", "cifa100.yaml"]
env:
- name: DECKARD_DEVICE
nvidia-tesla-p100
# - name: REDIS_HOST
# value: "redis"
# - name: REDIS_PORT
# value: "6379"
# - name: REDIS_DB
# value: "0"
# - name: REDIS_PASSWORD
# value: ""
resources:
limits:
nvidia.com/gpu: 1
volumeMounts:
- mountPath: /deckard/
name: mypvc
volumes:
- name: mypvc
persistentVolumeClaim:
claimName: podpvc
14 changes: 14 additions & 0 deletions examples/power/conf/deploy-nvidia-tesla-p100/default.yaml
Original file line number Diff line number Diff line change
@@ -0,0 +1,14 @@
num_nodes: 1
cluster_name: k8s-cluster
gpu_type: nvidia-tesla-v100
gpu_count: 1
gpu_driver_version: default
machine_type: n1-standard-2
min_nodes: 1
max_nodes: 1
storage_config: conf/deploy/sclass.yaml
persistent_volume_claim: conf/deploy/pvc.yaml
pod : conf/deploy/pod.yaml
image_project: ubuntu-os-cloud
image_family: ubuntu-2204-lts
mount_directory: /mnt/filestore
32 changes: 32 additions & 0 deletions examples/power/conf/deploy-nvidia-tesla-p100/mnist.yaml
Original file line number Diff line number Diff line change
@@ -0,0 +1,32 @@
apiVersion: v1
kind: Pod
metadata:
name: deckard
spec:
containers:
- name: deckard
image: ghcr.io/simplymathematics/deckard:main
imagePullPolicy: Always
workingDir: /deckard/examples/power/
args: ["python", "-m", "deckard", "--config_file", "mnist.yaml"]
env:
- name: DECKARD_DEVICE
nvidia-tesla-p100
# - name: REDIS_HOST
# value: "redis"
# - name: REDIS_PORT
# value: "6379"
# - name: REDIS_DB
# value: "0"
# - name: REDIS_PASSWORD
# value: ""
resources:
limits:
nvidia.com/gpu: 1
volumeMounts:
- mountPath: /deckard/
name: mypvc
volumes:
- name: mypvc
persistentVolumeClaim:
claimName: podpvc
11 changes: 11 additions & 0 deletions examples/power/conf/deploy-nvidia-tesla-p100/pvc.yaml
Original file line number Diff line number Diff line change
@@ -0,0 +1,11 @@
kind: PersistentVolumeClaim
apiVersion: v1
metadata:
name: podpvc
spec:
accessModes:
- ReadWriteMany
storageClassName: filestore-sc
resources:
requests:
storage: 256Gi
Loading
Loading