Manage the machine learning lifecycle with MLflow on a Google Cloud C4A Axion virtual machine
Introduction
Learn about MLflow and Google Axion C4A for machine learning
Configure Google Cloud firewall rules for MLflow
Create a Google Axion C4A virtual machine for MLflow
Install MLflow and track machine learning experiments
Deploy MLflow models as REST APIs
Next Steps

  Log an issue

  Fork and edit

  Discuss on Discord

About this Learning Path

Skill level:  Introductory
Reading time:  30 min
Last updated:  29 Sep 2026
Skill level:
  Introductory
Reading time:
  30 min
Last updated:
  29 Sep 2026
Author:Pareena Verma, Arm
Arm IP:Neoverse
Tags: ML Google Axion Linux MLflow Python scikit-learn
Author:
  • Pareena Verma, Arm
Arm IP:
Neoverse
Tags:
ML Google Axion Linux MLflow Python scikit-learn

Who is this for?

This is an introductory topic for DevOps engineers, ML engineers, and software developers who want to manage the machine learning lifecycle using MLflow on SUSE Linux Enterprise Server (SLES) Arm64, track experiments, version models, and deploy models as scalable APIs.

What will you learn?

After completing this Learning Path, you'll be able to perform the following tasks:

  • Install and configure MLflow on an Arm-based Google Cloud C4A virtual machine (VM).
  • Track experiments, log metrics, and compare runs using MLflow Tracking.
  • Manage and version models using the MLflow Model Registry.
  • Deploy models as APIs and validate end-to-end ML workflows.

Prerequisites

Before starting, you'll need the following prerequisites:

Summary

AI-assisted

This summary was drafted with an approved AI-assisted workflow and reviewed by Arm contributors before publication. Human technical review remains part of the process so the final page reflects engineering rigor, accuracy, and Arm editorial standards.

Close
?
You’ll manage the machine learning lifecycle with MLflow on a Google Axion C4A virtual machine. First, you’ll provision a SUSE Linux VM, configure firewall access, and install MLflow with Python 3.11. Then, you’ll start the tracking server, run experiments, and compare logged metrics. Finally, you’ll register model versions and aliases, start model serving, and validate the deployment through the MLflow interface and REST API predictions.

Frequently asked questions

AI-assisted

These FAQs were drafted with an approved AI-assisted workflow and reviewed by Arm contributors before publication. Human technical review remains part of the process so the final page reflects engineering rigor, accuracy, and Arm editorial standards.

Close
?
Which VM configuration should I select for this setup?
Select a Google Axion C4A instance with the c4a-standard-4 machine type, which provides four vCPUs and 16 GB of memory. Use SUSE Linux for the VM.
What firewall rules do I need before launching MLflow?
Allow inbound TCP access to port 5000 for the MLflow interface and port 6000 for the model-serving API. Attach the firewall rule to your VM through its network tag.
How do I know that the MLflow tracking server is running?
Check the startup log for the listening address, then open http://<VM-IP>:5000. You should see the MLflow interface after the server starts.
Why do I need to set MLFLOW_TRACKING_URI?
Set MLFLOW_TRACKING_URI to direct your MLflow client to the tracking server instead of a local directory. Run export MLFLOW_TRACKING_URI=http://127.0.0.1:5000.
How can I confirm that my model is registered and served correctly?
Confirm that your model appears in the MLflow Model Registry with a version and assigned alias. Then, call the REST endpoint and verify that its response contains predictions from the served model.
Next