Docs
PackagesRunpod v1.9.99 published on Wednesday, Jul 3, 2024 by Runpod
RunPod
The RunPod provider for Pulumi can be used to provision RunPod resources. The RunPod provider must be configured with RunPod’s API keys to deploy and update resources in RunPod.
Config
To begin with, please set your RunPod API key using Pulumi.
pulumi config set --secret runpod:token
Note
Please make sure that you are inside the Python virtual environment created by Pulumi when using the Python SDK.
Example
import * as pulumi from "@pulumi/pulumi";
import * as runpod from "@runpod-infra/pulumi";
const myTemplate = new runpod.Template("testTemplate", {
containerDiskInGb: 5,
dockerArgs: "python handler.py",
env: [
{
key: "key1",
value: "value1",
},
{
key: "key2",
value: "value2",
},
],
imageName: "runpod/serverless-hello-world:latest",
isServerless: true,
name: "Testing Pulumi V1",
readme: "## Hello, World!",
volumeInGb: 0,
});
const testNetworkStorage = new runpod.NetworkStorage("testNetworkStorage", {
name: "testStorage1",
size: 10,
dataCenterId: "US-OR-1",
});
const myRandomPod = new runpod.Pod("myRandomPod", {
cloudType: "ALL",
networkVolumeId: testNetworkStorage.networkStorage.apply(
// @ts-ignore
(networkStorage) => networkStorage.id
),
gpuCount: 1,
volumeInGb: 50,
containerDiskInGb: 50,
minVcpuCount: 2,
minMemoryInGb: 15,
gpuTypeId: "NVIDIA GeForce RTX 4090",
name: "RunPod Pytorch",
imageName: "runpod/pytorch:latest",
dockerArgs: "",
ports: "8888/http",
volumeMountPath: "/workspace",
env: [
{
key: "JUPYTER_PASSWORD",
value: "rns1hunbsstltcpad22d",
},
],
});
const myRandomEndpoint = new runpod.Endpoint("myRandomEndpoint", {
gpuIds: "AMPERE_16,AMPERE_24,-NVIDIA L4",
idleTimeout: 100,
locations: "CA-MTL-2,CA-MTL-3,EU-RO-1,US-CA-1,US-GA-1,US-KS-2,US-OR-1,CA-MTL-1,US-TX-3,EUR-IS-1,EUR-IS-2,SEA-SG-1",
name: "myRandomEndpoint",
networkVolumeId: testNetworkStorage.networkStorage.apply(
// @ts-ignore
(networkStorage) => networkStorage.id
),
scalerType: 'REQUEST_COUNT',
scalerValue: 2,
templateId: myTemplate.template.apply(t => t.id),
workersMax: 2,
workersMin: 1,
})
export const template = {
value: myTemplate.template,
};
export const endpoint = {
value: myRandomEndpoint.endpoint,
};
export const pod = {
value: myRandomPod.pod,
};
export const networkStorage = {
value: testNetworkStorage.networkStorage,
};
package main
import (
"github.com/pulumi/pulumi/sdk/v3/go/pulumi"
"github.com/runpod/pulumi-runpod-native/sdk/go/runpod"
)
func main() {
pulumi.Run(func(ctx *pulumi.Context) error {
testNetworkStorage, err := runpod.NewNetworkStorage(ctx, "testNetworkStorage", &runpod.NetworkStorageArgs{
Name: pulumi.String("testStorage1"),
Size: pulumi.Int(5),
DataCenterId: pulumi.String("EU-RO-1"),
})
if err != nil {
return err
}
myRandomPod, err := runpod.NewPod(ctx, "myRandomPod", &runpod.PodArgs{
CloudType: pulumi.String("ALL"),
NetworkVolumeId: testNetworkStorage.NetworkStorage.ApplyT(func(networkStorage runpod.NetworkStorageType) (*string, error) {
return &networkStorage.Id, nil
}).(pulumi.StringPtrOutput),
GpuCount: pulumi.Int(1),
VolumeInGb: pulumi.Int(50),
ContainerDiskInGb: pulumi.Int(50),
MinVcpuCount: pulumi.Int(2),
MinMemoryInGb: pulumi.Int(15),
GpuTypeId: pulumi.String("NVIDIA GeForce RTX 4090"),
Name: pulumi.String("RunPod Pytorch"),
ImageName: pulumi.String("runpod/pytorch"),
DockerArgs: pulumi.String(""),
Ports: pulumi.String("8888/http"),
VolumeMountPath: pulumi.String("/workspace"),
Env: runpod.PodEnvArray{
&runpod.PodEnvArgs{
Key: pulumi.String("JUPYTER_PASSWORD"),
Value: pulumi.String("rns1hunbsstltcpad22d"),
},
},
})
if err != nil {
return err
}
myTemplate, err := runpod.NewTemplate(ctx, "myTemplate", &runpod.TemplateArgs{
ContainerDiskInGb: pulumi.Int(5),
DockerArgs: pulumi.String("python3 -m http.server 8080"),
Env: runpod.PodEnvArray{
&runpod.PodEnvArgs{
Key: pulumi.String("JUPYTER_PASSWORD"),
Value: pulumi.String("rns1hunbsstltcpad22d"),
},
},
ImageName: pulumi.String("runpod/serverless-hello-world:latest"),
IsServerless: pulumi.Bool(true),
Name: pulumi.String("Testing Pulumi V1"),
Readme: pulumi.String("## Hello, World!"),
VolumeInGb: pulumi.Int(0),
})
if err != nil {
return err
}
myEndpoint, err := runpod.NewEndpoint(ctx, "myEndpoint", &runpod.EndpointArgs{
GpuIds: pulumi.String("AMPERE_16"),
Name: pulumi.String("Pulumi Endpoint Test V2 -fb"),
TemplateId: myTemplate.Template.Id(),
WorkersMax: pulumi.Int(2),
WorkersMin: pulumi.Int(1),
IdleTimeout: pulumi.Int(6),
Locations: pulumi.String("EU-RO-1"),
NetworkVolumeId: testNetworkStorage.NetworkStorage.Id(),
ScalerType: pulumi.String("QUEUE_DELAY"),
ScalerValue: pulumi.Int(4),
})
if err != nil {
return err
}
ctx.Export("pod", myRandomPod.Pod.Id())
ctx.Export("networkStorage", testNetworkStorage.NetworkStorage.Id())
ctx.Export("template", myTemplate.Template.Id())
ctx.Export("endpoint", myEndpoint.Endpoint.Id())
return nil
})
}
source venv/bin/activate
import pulumi
import runpodinfra as runpod
from loguru import logger
import json
def fetch_id(a):
if type(a) == runpod.outputs.NetworkStorage:
return a.id
else:
return a
def fetch_template_id(a):
if type(a) == runpod.outputs.Template:
return a.id
else:
return a
try:
test_network_storage = runpod.NetworkStorage("testNetworkStorage",
name="testStorage1",
size=5,
data_center_id="EU-RO-1")
my_random_pod = runpod.Pod("myRandomPod",
cloud_type="ALL",
network_volume_id=test_network_storage.network_storage.apply(lambda x : fetch_id(x)),
gpu_count=1,
volume_in_gb=50,
container_disk_in_gb=50,
min_vcpu_count=2,
min_memory_in_gb=15,
gpu_type_id="NVIDIA GeForce RTX 4090",
name="RunPod Pytorch",
image_name="runpod/pytorch",
docker_args="",
ports="8888/http",
volume_mount_path="/workspace",
env=[runpod.PodEnvArgs(
key="JUPYTER_PASSWORD",
value="rns1hunbsstltcpad22d",
)])
my_random_template = runpod.Template("myRandomTemplate",
container_disk_in_gb = 5,
container_registry_auth_id = "",
docker_args="python handler.py",
env=[{"key": "hi", "value": "hello"}],
image_name="runpod/serverless-hello-world:latest",
is_public=False,
is_serverless=True,
name="Generated Serverless Template",
ports="1293/http",
readme="Some readme", # pass some value to this. Won't work otherwise
start_jupyter=False,
start_ssh=False,
volume_in_gb=20,
volume_mount_path="/workspace",
)
my_random_endpoint = runpod.Endpoint("myRandomEndpoint",
gpu_ids="AMPERE_16,AMPERE_24,-NVIDIA L4",
idle_timeout=100,
locations="CA-MTL-2,CA-MTL-3,EU-RO-1,US-CA-1,US-GA-1,US-KS-2,US-OR-1,CA-MTL-1,US-TX-3,EUR-IS-1,EUR-IS-2,SEA-SG-1",
name="myRandomEndpoint",
network_volume_id=test_network_storage.network_storage.apply(lambda x : fetch_id(x)),
scaler_type='REQUEST_COUNT',
scaler_value=2,
template_id=my_random_template.template.apply(lambda x : fetch_template_id(x)),
workers_max=2,
workers_min=1,
)
pulumi.export("pod", {
"value": my_random_pod.pod,
})
pulumi.export("networkStorage", {
"value": test_network_storage.network_storage,
})
pulumi.export("testTemplate", {
"value": my_random_template.template,
})
pulumi.export("testEndpoint", {
"value": my_random_endpoint.endpoint,
})
except Exception as e:
logger.exception(e)
name: provider-runpod-native
runtime: yaml
plugins:
providers:
- name: runpod
path: ../../bin
resources:
testNetworkStorage:
type: runpod:NetworkStorage
properties:
name: testStorage1
size: 5
dataCenterId: EU-RO-1
myRandomPod:
type: runpod:Pod
properties:
cloudType: ALL
networkVolumeId: ${testNetworkStorage.networkStorage.id}
gpuCount: 1
volumeInGb: 60
containerDiskInGb: 50
minVcpuCount: 2
minMemoryInGb: 15
gpuTypeId: "NVIDIA GeForce RTX 4090"
name: "RunPod Pytorch"
imageName: "runpod/pytorch"
dockerArgs: ""
ports: "8888/http"
volumeMountPath: "/workspace"
env:
- key: "JUPYTER_PASSWORD"
value: "rns1hunbsstltcpad22d"
myRandomTemplate:
type: runpod:Template
properties:
containerDiskInGb: 20
containerRegistryAuthId: ""
dockerArgs: "python3 -m http.server 8080"
env: [{ key: "JUPYTER_PASSWORD", value: "rns1hunbsstltcpad22d" }]
imageName: "nginx:latest"
isPublic: false
isServerless: true
name: "RunPod Nginx"
ports: "8080/http"
readme: "Test template"
startJupyter: false
startSsh: false
volumeInGb: 10
volumeMountPath: "/workspace"
myRandomEndpoint:
type: runpod:Endpoint
properties:
gpuIds: "AMPERE_16,AMPERE_24,-NVIDIA L4"
idleTimeout: 100
locations: "CA-MTL-2,CA-MTL-3,EU-RO-1,US-CA-1,US-GA-1,US-KS-2,US-OR-1,CA-MTL-1,US-TX-3,EUR-IS-1,EUR-IS-2,SEA-SG-1"
name: "myRandomEndpoint"
networkVolumeId: ${testNetworkStorage.networkStorage.id}
scalerType: "REQUEST_COUNT"
scalerValue: 2
templateId: ${myRandomTemplate.template.id}
workersMax: 2
workersMin: 1
outputs:
pod:
value: ${myRandomPod.pod}
networkStorage:
value: ${testNetworkStorage.networkStorage}
template:
value: ${myRandomTemplate.template}
endpoint:
value: ${myRandomEndpoint.endpoint}