We will take a ground-up approach; in this way, we'll learn more, which, as a developer, is always helpful at the time of debugging. We will use Visual Studio Community edition (Free) with Docker support tools installed separately, and will not use the extra features of VS 2017.

Our frontend microservice is a simple single-page static (HTML + JS) application, which we've decided to host on the NGINX web server in a Linux-based container.

Our backend microservice is a simple REST API application, which is based on ASP.NET Core hosted on the Kestrel HTTP web server inside a Linux-based container.

For a richer experience, we will install both containers on a CentOS Linux, and test the Math App by interacting with the frontend SPA from a browser in Windows. Once we have achieved this successfully, we will push our docker container images to Docker Hub, the public docker images repository. We will pull these images, and then execute them inside our containers created using the ACS.

Following this approach, you'll not only understand the implementation of our architecture better, but also get the more fine-grained process of implementing the multi-container-based microservices architecture, which you would not grasp completely if you followed a VS 2017-based docker container implementation as your first application. In addition to that, it makes you cloud agnostic; you will be able to deploy container-based applications, locally on Dev machine(s), on-premises servers, or other popular cloud infrastructure providers.