Building Intelligent Cloud Applications

Chapter 6. Deployment and Scaling

After you develop a first version of any application, you need to deploy it to the infrastructure on which it will run.

By their very nature, serverless applications do not have an issue with scaling. But it is also true that an event-driven serverless project usually relies on many services that your cloud environment provides. For cases in which you need a more tailored setup, Kubernetes gives you a way to configure your customized containers to your needs. In this chapter, we explain how to use Azure Kubernetes Service (AKS) to orchestrate your custom containers.

Deployment Options

Deploying your Function App from the command line, as we have done so far, is just the first way to put your development on the Azure infrastructure. Sadly, as Python language support is in preview as of this writing, we must rely on this single method for deployment: a custom container to compile your Python code with all the virtual environment requirements is needed. But for a Function App in any other programming language, we have multiple options that we explain here so that you can take advantage of the concepts as soon as they become available for Python.

On your Function App web page in the Azure Portal, where all the contents and options are displayed, click the Platform Features tab. Then, in the “Code deployment” section, click the “Deployment center” link. When you visit it, several source control options are available, including a number of locations ...