MLflow on Aruba Cloud
Deploy MLflow — an open-source platform for managing the end-to-end ML lifecycle, including experiment tracking, model registry, and artifact storage — on Aruba Cloud using Terraform and cloud-init. Uses Managed MySQL 8.0 as the tracking backend and nginx with HTTP Basic Auth as a reverse proxy.
Provider version: arubacloud/arubacloud
~> 1.0| Terraform: ≥ 1.9
Introduction
This example deploys:
- MLflow tracking server (pip, native install) on a CSO2A4 VM
- Managed MySQL 8.0 (Aruba Cloud DBaaS) for experiment and run metadata
- nginx reverse proxy with HTTP Basic Auth on port 80
- Artifact files (models, plots) stored locally on the VM disk
- Optional HTTPS via Actalis ACME Certificates or Let's Encrypt when a custom domain is provided
Architecture Overview
Variables
Required
| Variable | Description |
|---|---|
arubacloud_client_id | ArubaCloud OAuth2 client ID |
arubacloud_client_secret | ArubaCloud OAuth2 client secret |
Optional
| Variable | Default | Description |
|---|---|---|
app_name | "mlflow" | Short name used in resource names (2–8 chars) |
environment | "prod" | Environment label |
location | "ITBG-Bergamo" | ArubaCloud region |
zone | "ITBG-1" | Availability zone |
billing_period | "Hour" | "Hour" or "Month" |
vm_flavor | "CSO2A4" | CloudServer flavor (2 vCPU / 4 GB) |
vm_disk_size_gb | 40 | Boot disk size — increase for large artifact storage |
ssh_public_key | project owner key | SSH public key content |
ssh_cidr | "0.0.0.0/0" | CIDR for SSH access |
dbaas_flavor | "DBO2A8" | Managed MySQL flavor |
db_storage_gb | 20 | DBaaS initial storage in GB |
db_password | pre-set default | MySQL password (min 8 chars, no newlines) |
mlflow_admin_user | "admin" | Username for HTTP Basic Auth |
mlflow_admin_password | pre-set default | Password for HTTP Basic Auth (min 8 chars) |
mlflow_version | "2.16.0" | MLflow version to install |
domain | "" | Custom domain for HTTPS (e.g. mlflow.example.com); leave empty for HTTP over Elastic IP |
acme_eab_kid | "" | Actalis ACME EAB key ID; leave empty to fall back to Let's Encrypt |
acme_eab_hmac_key | "" | Actalis ACME EAB HMAC key; required when acme_eab_kid is set |
Outputs
| Output | Description |
|---|---|
app_url | MLflow tracking server URL |
mlflow_tracking_uri | MLFLOW_TRACKING_URI to set in experiment code |
vm_public_ip | Public IP of the VM |
ssh_command | SSH command to connect |
Deployment Instructions
1. Clone and navigate
git clone https://github.com/arubacloud/terraform-arubacloud-examples.git
cd terraform-arubacloud-examples/mlflow
2. Configure variables
cp terraform.tfvars.example terraform.tfvars
3. Deploy
terraform init
terraform plan
terraform apply
Bootstrap takes approximately 5–10 minutes.
4. Connect your experiments
import mlflow
mlflow.set_tracking_uri("http://<vm_public_ip>") # or the domain
mlflow.set_experiment("my-experiment")
with mlflow.start_run():
mlflow.log_param("alpha", 0.5)
mlflow.log_metric("rmse", 0.82)
Set HTTP Basic Auth credentials in your environment:
export MLFLOW_TRACKING_USERNAME=admin
export MLFLOW_TRACKING_PASSWORD=<mlflow_admin_password>