How MSPs can Deliver Managed Docker Environments with Containers-as-a-Service
Brought to you by MSPmentor
Containers-as-a-Service, or CaaS, is becoming one of the most popular ways for companies to use Docker. Here's what MSPs need to know about adding CaaS to the services they deliver to customers.
CaaS, which should not be confused with Compliance-as-a-Service, provides a turn-key solution for organizations to run Docker containers. CaaS platforms come prebuilt with all of the components required to set up, orchestrate, manage and monitor containerized applications.
For companies looking to migrate workloads from legacy infrastructure to Dockerized infrastructure, CaaS is the way to go.
CaaS and MSPs
If you're an MSP seeking to provide CaaS to your customers, you should know that CaaS offerings fall into two main categories.
The first is CaaS services that run in the public cloud, such as Amazon ECS and Azure Container Service. These solutions will be less useful for MSPs, since there is little value that MSPs can add to container services that are already managed and running on public cloud infrastructure.
The second, more interesting type of CaaS platforms are those that MSPs can install on any on-premise or cloud-based server. Most of these CaaSes can be easily installed on any physical or virtual server running a modern version of Linux. Examples include Rancher, Kontena and OpenShift.
The latter type of CaaS offerings can help MSPs to build a managed container service for customers. MSPs add value by setting up the CaaS, providing infrastructure to run it on and supporting it. If you can beat the rates charged by public-cloud CaaSes like ECS and Azure Container Service, or deliver better support and customization, your CaaS offering will provide superior value to your customers.
So, if you'd like to add managed Docker services to the offerings you provide to customers, consider setting up a CaaS platform and using it to deliver Docker hosting. The CaaS will save you the trouble of building a complete Docker stack from scratch and keep the learning curve manageable.