Containers are processes that run on a host. This seems very simple, but how will this work on a pool of nodes? If we are looking for high availability, being able to run our containers on any host from a pool will ensure execution everywhere. But this approach requires some special logic in our applications. Our applications must be completely portable and avoid friction and dependencies on any host. Applications with many dependencies are always less portable. We need to find a way to manage status data for containers. We will review different persistence strategies in this chapter.
On the other hand, the aforementioned pool of hosts must be able to communicate with all containers. In this chapter, ...
