Containers have gained a lot of ground with digital enterprises, no doubt. According to a recent report, containers have become key enablers as organizations adopt continuous delivery as part of their digital transformation. Early adopters who went through the learning cycle are now looking to ‘operationalize’ them at a broader scale within the engineering organizations.
There are interesting discussions within the DevOps buying communities to scale the deployment of these applications using containers more easily and efficiently. We should not forget that early adopters have worked on the bleeding edge of this technology, and there’s a lot of learning from those initial experiments.
From a survey conducted by CloudFoundry, container management has been one of the top challenges for enterprises who have adopted containers. An even steeper challenge than monitoring, storage, or security/isolation. The biggest deployment worry has been to sieve through the complexity to integrate this new technology into legacy environments.
There is no denying that operationalizing container technology is quite complex, and as you move beyond the early adopter stage and start setting up a path to scale, there are several key factors to consider. Here, I will highlight a few factors that are often overlooked, but ones we have found to be important, based on engagements with several customers across different verticals such as technology, manufacturing, finance and retail.
- Consider a container-native solution: There are several container solutions out there, including ones from traditional enterprise players such as IBM and HP. But one key thing to keep in mind here is to evaluate container native solutions. Here's a great blog here defining container-native. Container native treats containers as the first-class unit of infrastructure, and since containers can impact many existing IT processes this becomes an important distinction. And ideally, it is software that ‘just does not happen’ to work on containers, but is itself built using containers, for containers! There’s a huge amount of science and best practice behind it. When you go with the ‘native’ guys you know they have deep understanding of what it takes to use containers in production.
- Consider a cloud-agnostic solution: As an enterprise, it is essential for you to think cloud agnostic when you’re considering containers or any such related technology. Reason – although you might be running on a specific cloud service today, your business model and hence your application should be flexible enough to use multi-cloud or a completely different cloud solution when required.
- Consider an “as a service” delivery model: There are quite a few container deployment solutions out there but most of them will require your DevOps teams to do the heavy lifting, taking away precious resources from your business deliverables. But what if you had container management delivered with a SaaS approach? A cloud service that would help you to not only deploy your applications in containers but also manage and monitor them real time and provide all the learning and best practices around it. This should include 24×7 support, well-defined availability, and security from a container savvy team.
To sum up: Containers alone are not enough!
So, it is clear that as container adoption happens, enterprises will want to scale efforts. Just having container technology will not be enough and there will be a need to manage them with a cloud application platform.