The tools for managing the various containers that comprise distributed cloud-based applications.
A container orchestrator ensures that when launching a distributed cloud application, all the components needed to run that application are available. Its role is much like that of an orchestra’s conductor. Kubernetes is one the most popular container orchestrators.
If you’re looking to take full advantage of the flexibility and scalability of cloud computing, you will want to consider a container orchestrator.
What is it?
Modern distributed cloud applications are complex. Container orchestrators ensure everything needed to run an application is there. The orchestrator ensures that applications keep running, even if individual containers go down, and they enable capacity to be added on demand.
Container orchestration is useful when implementing a microservices architecture, as they enable all the necessary components, such as network and storage, to be managed when delivered as containers.
What’s in for you?
Container orchestrators ease the deployment and management of containers — that can mean reduced time to market, as your software delivery cycle shortens. If you’re building distributed cloud applications that use containers — as part of becoming a modern digital business — the cloud orchestrator will be an important part of that.
Your container orchestrator can enable you to gain control of cloud spend, by implementing rules on the amount of compute power or storage that your applications are given.
What are the trade offs?
The container orchestration space is evolving rapidly. That means there’s a training overhead to consider if you choose to run the orchestrator yourself.
Your cloud service providers can manage the orchestrator for you. But if you do choose this option, be warned that each provider has a unique take on how to do this — for instance, they may each treat storage differently. That means it’s harder to port you applications between cloud providers.
Some parts of your IT infrastructure may not work well when using container orchestration. For instance, some databases aren’t built to support orchestration — they were built for fixed data center infrastructure and don't work well in cloud environments where instances can be relocated. Building a highly available service on top of such databases is not trivial.