Deploy a Container Group with the Public API

Prerequisites

Before you begin, make sure you have the following:

  • An account with Salad Cloud. If you don't have one yet, you can sign up for a free trial at https://portal.salad.com .
  • A project in Salad Cloud. If you don't have a project yet, you can create one by following the instructions in the Salad Cloud documentation.
  • A container image that you want to deploy. If you don't have a container image yet, you can build one using a tool like Docker.

Step 1: Obtain Your API Key

To authenticate your API requests, you'll need an API key. To obtain your API key, follow these steps:

  • Log in to your Salad Cloud account at https://portal.salad.com
  • Click on your profile in the top right corner of the page, then click API Keys.
  • Copy the API key that is displayed. You'll need this in the next step.

Step 2: Deploy the Container Image

To deploy a container group using the Salad Cloud Public API, you'll need to send a POST request with specific parameters inside the data field along with headers. Here's how to do it:

Send a POST request to the following URL:

https://api.salad.com/api/public/organizations/{organization_name}/projects/{project_name}/containers

Replace {organization_name} and {project_name} with the name of your organization and project, respectively.

Include the following headers in your request:

  • Salad-Api-Key: {API_KEY}
  • accept: application/json
  • content-type: application/json

Include a JSON payload in the data field of the request body with the necessary configuration parameters. Here's an example using curl:

curl --request POST \
     --url https://api.salad.com/api/public/organizations/{organization_name}/projects/{project_name}/containers \
     --header 'Salad-Api-Key: {API_KEY}' \
     --header 'accept: application/json' \
     --header 'content-type: application/json' \
     --data '{
       "container": {
         "resources": {
           "memory": 2048, 
           "cpu": 1,
           "gpu_classes": [
             "ed563892-aacd-40f5-80b7-90c9be6c759b", // ID for 3090s
             "a5db5c50-cbcb-4596-ae80-6a0c8090d80f", // ID for 3090Tis
             "9998fe42-04a5-4807-b3a5-849943f16c38" // ID for 4090s
           ],
           "image": "{CONTAINER_IMAGE_URL}" // Container image URL
         },
         "restart_policy": "always",
         "country_codes": ["af"],
         "networking": {
           "protocol": "http",
           "auth": true,
           "port": 8080
         },
         "name": "demo-test",
         "replicas": 1  // Number of replicas
       }
     }'

In the data field, you can configure various parameters to define your container group's behavior and characteristics. For detailed information on these parameters and their options, please refer to the Create a Container Group

Step 3: Select Multiple GPUs

To enable the use of multiple GPUs within a container, you'll need to:

  1. Retrieve GPU IDs : Begin by obtaining the list of available GPUs. To do this, execute the following curl command

    curl --request GET \
     --url https://api.salad.com/api/public/organizations/organization-name/gpu-classes \
     --header 'Salad-Api-Key: {api-key}' \
     --header 'accept: application/json'

  2. Note GPU IDs : Once you receive the response, identify and make note of the unique IDs corresponding to the GPUs you intend to use.

  3. Configure Your Workload : When setting up your workload, include the GPU IDs in the gpu_classes field. For example, if you plan to utilize GPUs like 3090s, 3090Tis, and 4090s, structure your configuration like this

    "container": {
  "resources": {
    "cpu": 1,
    "memory": 1024,
    "gpu_classes": [
      "ed563892-aacd-example-90c9be6c759b",  // ID for 3090s
      "a5db5c50-cbcb-example-6a0c8090d80f",  // ID for 3090Tis
      "9998fe42-04a5-example-849943f16c38"  // ID for 4090s
    ]
  }
}

Note: you can also test it on our API reference page

Step 4: Check the Deployment Status

After you send the POST request in previous step, Salad Cloud will begin the process of deploying your container image. To check the status of the deployment, you can send a GET request to the following URL.

https://api.salad.com/api/public/organizations/{organization_name}/projects/{project_name}/containers/{container_group_name}

Replace {organization_name}, {project_name}, and {container_group_name} with the name of your organization, project, and the container group you want to check, respectively.

Include the following headers in your GET request:

  • Salad-Api-Key: {API_KEY}
  • accept: application/json

Here's an example of how to send the GET request using curl:

curl --request GET \
     --url https://api.salad.com/api/public/organizations/{organization_name}/projects/{project_name}/containers/{container_group_name} \
     --header 'Salad-Api-Key: {API_KEY}' \
     --header 'accept: application/json'

The response will include the current status of the deployment, as well as other details about the container.

Note: By default, when you deploy a container group via the public API, auto-start is set to false. You can start it by either visiting the portal or using the public API itself

Updated 6 months ago