Containers in the Cloud
Learn about cloud computing with containers and some points that are of concern to deployment environments.
We'll cover the following
How containers run in cloud
Containers on cloud VMs combine the challenges of both containers and the cloud. The containers have short-lived, ephemeral identities. Connecting them means linking ports across different VMs, possibly in different zones or regions. Designing individual services to run in this kind of deployment is not that much different from designing them to run in containers in the data center. Most of the big challenges arise from building those containers into a whole system.
In a sense, using containers pushes some complexity out of the boxes and into the control plane. We’ll look at the control plane in the Control Plane chapter.
Wrapping up
The range of deployment environments has widened thanks to cloud computing and platform-as-a-service offers. These environments move the boundary of responsibility back and forth between application development, platform development, operations, and infrastructure.
Despite that, some considerations are common to every kind of environment:
- How is the network structured? Is there just one or are there several? Will a machine have NICs on different networks with different jobs?
- Do machines have long-lasting identities?
- Are machines automatically set up and torn down? If so, how do we manage the images for them?
Finding or building the answer to these questions never appears on a Kanban board or a Jira ticket, but they’re essential to making a smooth transition to operations. Given a stable foundation to build upon, we need to look at how individual machine instances in that environment will behave and how we will control them. We’ll look at those issues in the next chapter.
Create a free account to view this lesson.
By signing up, you agree to Educative's Terms of Service and Privacy Policy