Spice.ai Enterprise (BYOL)

Ensure the following tools and resources are ready before starting:

• Docker: Install from https://docs.docker.com/get-docker/ .
• AWS CLI: Install from https://docs.aws.amazon.com/cli/latest/userguide/getting-started-install.html .
• AWS ECR Access: Authenticate to the AWS Marketplace registry: aws ecr get-login-password --region us-east-1 | docker login --username AWS --password-stdin 709825985650.dkr.ecr.us-east-1.amazonaws.com
• Spicepod Configuration: Prepare a spicepod.yaml file in your working directory. A spicepod is a YAML manifest file that configures which components (i.e. datasets) are loaded. Refer to https://spiceai.org/docs/getting-started/spicepods  for details.
• AWS ECS Prerequisites (for ECS deployment): An ECS cluster (Fargate or EC2) configured in your AWS account. An IAM role for ECS task execution (e.g., ecsTaskExecutionRole) with permissions for ECR, CloudWatch, and other required services. A VPC with subnets and a security group allowing inbound traffic on ports 8090 (HTTP) and 50051 (Flight).

Running the Container

1. Ensure the spicepod.yaml is in the current directory (e.g., ./spicepod.yaml).
2. Launch the container, mounting the current directory to /app and exposing HTTP and Flight endpoints externally:

docker run --name spiceai-enterprise
-v $(pwd):/app
-p 50051:50051
-p 8090:8090
709825985650.dkr.ecr.us-east-1.amazonaws.com/spice-ai/spiceai-enterprise-byol:1.10.4-enterprise-models
--http 0.0.0.0:8090
--flight 0.0.0.0:50051

• The -v $(pwd):/app mounts the current directory to /app, where spicepod.yaml is expected.
• The --http and --flight flags set endpoints to listen on 0.0.0.0, allowing external access (default is 127.0.0.1).
• Ports 8090 (HTTP) and 50051 (Flight) are mapped for external access.

Verify and Monitor the Container

1. Confirm the container is running:

docker ps

Look for spiceai-enterprise with a STATUS of Up.

2. Inspect logs for troubleshooting:

docker logs spiceai-enterprise

Deploying to AWS ECS Create an ECS Task Definition and use this value for the image: 709825985650.dkr.ecr.us-east-1.amazonaws.com/spice-ai/spiceai-enterprise-byol:1.10.4-enterprise-models. Configure the port mappings for the HTTP and Flight ports (i.e. 8090 and 50051).

Override the command to expose the HTTP and Flight ports publically and link to the Spicepod configuration hosted on S3:

"command": [ "--http", "0.0.0.0:8090", "--flight", "0.0.0.0:50051", "s3://your_bucket/path/to/spicepod.yaml" ]

Register the task definition in your AWS account, i.e. aws ecs register-task-definition --cli-input-json file://spiceai-task-definition.json --region us-east-1

Then run the task as you normally would in ECS.