What you'll learn
- How to Deploy your Training Code as a Job via Truefoundry User Interface.
- The basic configurations you can have for your Job Deployments.
After you complete the guide, you will have successfully deployed a job to train a model on the iris dataset. Your jobs deployment dashboard will look similar to this:
To begin, log in to your Truefoundry account using your credentials. Upon successful login, you will be directed to the Deployments Dashboard for services.
- Navigate to the "Test out Service Deployment" section on the Deployments Dashboard. If there are existing deployments, you may need to scroll down slightly to find it.
- Click on the "Deploy using UI" button to initiate the deployment process.
A modal will appear with a "Where would you like to deploy?" option. Click on the "Search Workspace" bar, and select the desired workspace for your deployment.
If you don't have any workspace right now, you can create a new workspace by clicking the Create New Workspace link and following along this Documentation or contact your cluster admin in case of some issues
Once selected click on the "Next Step" button to continue with the deployment.
Now, you'll come across a deployment form featuring a range of configuration choices. These choices play a significant role in how your job gets deployed.
To make your first deployment smoother (thanks to the "Deploy using UI" button), we've already taken care of the basic settings. Nevertheless, it's important to know that you might want to fine-tune these options to suit your needs.
In the next section we are giving a very brief overview of what the options mean:
The Name field allows you to assign a unique identifier to the service within your workspace.
- Build Source (Deploy a Docker Image or Build and Deploy from Source Code):
The Build Source option lets you specify whether you want to deploy a pre-built Docker image or build and deploy from your source code.
- Build Specification (Build Using Dockerfile or Using Buildpack):
The Build Specification helps determine the method you want to use for building your service. You can choose to build using a Dockerfile, providing custom configuration for your environment, or you can opt for a Buildpack, which automates the build process based on your application's requirements.
- Environment Variables:
Environment Variables allow you to configure runtime settings for your service. You can provide key-value pairs that influence how your service behaves when it's running, such as database connection strings, API keys, or other configuration parameters.
The CPU resource allocation determines how much processing power your service can utilize. You can specify the desired amount of CPU capacity that your service can consume to perform its tasks efficiently.
Memory allocation dictates the amount of RAM your service can use. Define the memory limit to ensure that your service has enough memory available for its operations without causing performance issues.
Storage allocation refers to the amount of disk space your service can access. Specify the storage limit to ensure that your service has sufficient space to store files, logs, and other data it generates during its operation.
Specifies the maximum number of attempts to retry a job before it is marked as failed.
Now that all the deployment options are filled, you can proceed by clicking the "Create" button. This will initiate your deployment process. After clicking "Create," your Job Dashboard will resemble the following:
While your deployment is in progress, you can hover over the spinner icon to check the status of the deployment. After a brief moment, the deployment should become active, and your Job Dashboard will transform to appear as follows:
With the deployment now active, you can proceed to click on your specific job. Doing so will open up the dashboard dedicated to your job, where you'll have access to various details.
This will include the Run Job button.
Clicking on the following button will make an modal open, further submitting the modal with the Run Job button will redirect you to the Job Runs tab. After a few seconds of waiting, you should see your job showing Finished status.
Now click on the logs button to see your Jobs results.
Updated about 1 month ago