Containers are popular among the technology elite such as the leaders of hot startups and cloud providers. Enterprises don’t fall into that category. They manufacture tires, run banks or airlines, or operate another traditional businesses, so their needs may or may not be aligned with the vision of Silicon Valley.
In other words, they have to find the value of container technology for their situation, and they won’t move in that direction simply because it’s cool.
To get beyond cool, let me share the good, the bad, and the ugly about containers in the context of enterprise adoption.
Containers: What’s good for enterprises
Containers work well as an architecture. The concept of containers is not original, so it’s understood in the enterprise—even if under other labels. The technology is also solid. For example, the new Docker instances are well designed, using a centralized repository, and they can scale thanks to cluster managers such as Docker Swarm and Google Kubernetes. All major public clouds support containers, including AWS, Google, IBM, and Microsoft.
As a result, most promises made by container proponents are met. Moreover, if you use containers, they will lead you down the path to sound design of distributed, portable applications. What’s not to like?
Containers: What’s bad for enterprises
Containers have not been that successful when used for older applications. Although they’re an easy fit for new apps, they are too complex for older applications not designed from the ground up for containers.
The cost of moving existing applications to containers is more than proponents originally imagined. You have to redesign the applications to make the most of the container architecture, and that means more money, more time, and more risk. Thus, enterprises shy away from containerizing older applications.
Containers: What’s ugly for enterprises
What’s ugly about containers is the confusion that the ignorance about them creates. Rarely do I run into enterprises that understand the value of containers. For example, there’s a lot of discussion about choosing between containers and virtualization, but that debate has nothing to do with the real value of this technology.
The only way to combat the ignorance is education. But the hype leads the day, not education, so confusion will be with us until IT proactively educates itself rather than rely on the marketing by proponents and vendors. There is thoughtful, useful marketing available, but it’s hard to identify them until you understand the underlying issues of containers.
The confusion around containers is no different than the confusion around any new technology. But it’s different from many previous trumpeted technologies in that containers are quite useful and worth implementing.
The fact that container technologies have come at the same time that enterprises are moving applications to the cloud—and thus requiring decisions regarding what to do with older applications—could be the perfect storm of technology value. Done naively, however, it could be the perfect storm of misapplied technologies.