Last week Protocol hosted a panel discussion called So you want to go Multicloud. Now what?
There was one common theme that ran counter to the initial topic, and it started
from Priyanka Sharma of CNCF: you don’t decide to go multi-cloud. multi-cloud will find you.
That one comment shatters perhaps the biggest held belief: that companies choose to go multi-cloud at a strategic level. However the reality is far from the truth. There are a couple of perspectives to keep into account: for most companies, unless you’re running a SaaS company based mainly in Silicon Valley, you already have at least one data center running the applications critical to your basis. When making the formal decision to start going with a cloud provider, one of the major concerns you’ll have is how to integrate it and/or interoperate with your current data center. If you’re running vSphere and/or a Kubernetes cluster, then I believe you’re already there.
There is also the other perspective, and while the concept has been around for a very long time, the term that has grown in pervalence since the start of the pandemic is the Shadow IT. All of the cloud providers have made it very easy to sign up for an account, and most provide some kind of free credits or unlimited
functionality for the first month. For the IT professionals that have found greater flexibility at home and the need to get their projects done by any means necessary. This could start out with an engineer experimenting with the technology, or a decision made at the project level. In any case they made the
choice of technology, and you as the CIO are now running in the new multi-cloud world.
It is a combination of both perspectives that led to the revelation that the comment in going multi-cloud isn’t so much of a deliberate decision, but one that you will find yourself in whether you want to or not. The key point, and is the more appropriate question for you is, how intentional do you want to your strategy to look like going forward?
Especially since you will be weighing not just the costs of supporting the multiple environments and the regulatory hurdles, but also the exorbitant costs of transferring data and traffic into and out of the various cloud providers.
Note that there is a third perspective that I didn’t cover and that is integrating a merger or acquisition. To be honest, I don’t see this as a true integration though as it would need to be integrated regardless of whether it is on-prem, or a cloud provider.
At the end of the day, once you know where you’re at, then you can be more deliberate on where you want to go. And I suspect, if you’re thinking about going multi-cloud, the decision has already been made and you’re in the thick of it. The question then becomes how to manage the complexity, and what kind of controls you want to put into place going forward. Also, whether you’d like to include other possible providers to take advantage of their special deals for the time.