Building Intelligent Cloud Applications

Chapter 3. Serverless Application Programming Interfaces in Microsoft Azure

As we noted in Chapters 1 and 2, a serverless environment might be as beneficial for the cloud provider as it is for the developer. It is good for the provider because it is a way to ensure that its machines are always working on real tasks and not sitting idle, and it’s good for developers because they don’t need to think about server management. Now we introduce another concept, the idea of the application programming interface (API) as a way to access the many objects, methods, and functions available on the Azure platform.

APIs in Serverless Platforms

An API is how one program communicates with other programs. If you are going to build serverless machine learning solutions, cloud providers can accommodate your code in any of several programming languages. But you also must consider what serverless APIs these providers offer.

When you deploy your serverless code, it must interface with other services to do meaningful work. You will need, for example, to read and write data from databases and filesystems. But there are also many other services at your disposal, from pretrained machine learning modules to services for ingesting data from Internet of Things (IoT) devices, coordinating and communicating events, managing security, and exposing your own APIs.

The first step is to choose one serverless cloud provider and explore the different serverless APIs that it offers. There are several cloud providers ...