Using the Python Managed VM Environment

The managed VM hosting environment lets you run App Engine applications on configurable Compute Engine Virtual Machines (VMs) that are managed by Google. The VM hosting environment offers flexibility and additional CPU and memory options compared to the original App Engine sandboxed hosting environment.

The Python runtime is available as a Managed VM standard runtime. A standard runtime supports a subset of the App Engine service APIs. Currently, Managed VMs support these services:

• Datastore
• Logging
• Memcache
• Search
• Task Queue
• URL Fetch
• Users

If you choose to migrate an existing application to a managed VM, or to build a new one,be sure that the App Engine services that you need are available.

Configuring an app for managed VMs

You can switch your app to Managed VMs by changing just a few lines in your application's app.yaml file:

the runtime, vm, and api_version keys are required:

runtime: python27
vm: true
api_version: 1

You must also take into account these differences when migrating an existing app:

• Managed VMs support manual and automatic scaling. If your app uses basic scaling, you'll need to change that.
• Managed VMs do not assign instance classes. Instead, you select resources in the app.yaml file.

All the settings for managed VMs are described in theconfiguration instructions.

Testing and deploying managed VMs

In order to deploy a managed VM app into the Cloud, you must install the Google Cloud SDK. You'll also need to use the Google Developers Console to create a project and enable billing. The setup steps are explained in the Python Hello World starter project.

These additional sample projects are available:

• SQLite guestbook
• Zombie Apocalypse Simulator