Some time in the past I declared Kubernetes the winner within the container orchestration wars. Although I really like being proper, many within the cloud trade have framed Kubernetes because the end-all know-how that solves all our issues.
Thus we now have the overuse of Kubernetes to resolve all safety points, all infrastructure points, and even be a whole technique for know-how firms in search of the subsequent lily pad to leap onto. It’s all Kubernetes, on a regular basis.
As a practitioner of cloud computing, and somebody who leverages Kubernetes on-premises and within the public cloud, I can inform you that a lot of what’s good about Kubernetes is true. However, I can even inform you that Kubernetes shouldn’t be being thought of to assist sort out the core challenge going through us in 2020—cloud complexity.
There are two principal causes of cloud complexity:
First, the overuse of heterogeneity when deciding on cloud platforms. Although multicloud is a good suggestion, mixing hundreds of native APIs by an element of two or three right into a single unified platform makes the job of builders and the job of operations that rather more difficult.
Second, deploying a cloud answer with out correct planning. Deploying to a multicloud answer with minimized danger requires a minimum of some understanding of the place you at the moment are, the place you’re going, and tips on how to get there. Most enterprises can’t reply these questions and as a substitute proceed to function in a reactionary state.
There are additionally two options to cloud complexity:
First, abstraction. Using an abstraction layer with a least frequent denominator method removes you from the complexity of dealing straight with cloud-native instruments and interfaces.
Second, automation. Automating the usage of interfaces could make operations that a lot simpler and thus much less complicated.
Kubernetes tackles the automation challenge. The Kubernetes ecosystem, together with the current launch of Anthos, is about abstracting the purposes and information inside containers. The actual cash is in automating these containers in methods which are extremely scalable and cut back complexity on the similar time.
What considerations me is that individuals who should already take care of complexity don’t find out about automation or how Kubernetes can resolve these issues. They are targeted on particular proprietary instruments relatively than the concept of Kubernetes, which is extra of a metaconcept than a set of instruments.
If you’re coping with cloud complexity it’s essential give attention to the worth of automation, particularly, rising enabling applied sciences resembling Kubernetes. Kubernetes shouldn’t be a cure-all for cloud complexity, however it’s a idea that may be decomposed to particular helpful instruments and approaches.
Now, what about that lack of planning?
