Launching Serverless Tasks

ALB Mode

• Navigate to Compute → Elastic Deployment in the left menu.

• Click the Deploy button in the top-right corner to start the deployment wizard.

• In the configuration form:

  • Fill in all required fields (marked with *)

  • Optionally set additional parameters for custom needs

  • Click Add GPU to select the GPU type for your workload

You should put your service port in the HTTP Port field. The worker will be in Running status only when the port is ready.

Recommendation: choose GPUs from multiple regions to maximize redundancy and disaster recovery

• Click Deploy to launch your serverless task.

Queue Mode

You should put the port used for your service health check in the HTTP Port field. The worker will be in Running status only when the port is ready.

Custom Mode

Similarly to the ALB mode and Queue mode, you can choose Custom mode to hook up your serverless task with your own infrastructure (e.g. API Gateway, Queue)