One ring to rule them all: Best practices for a multi-cloud strategy


This blog was originally published by Concerto Cloud Services. Since then, Concerto Cloud Services has become DXC Concerto, the mid-market cloud offering within DXC Technology.

In terms of the future of cloud and best practices for its management, I recommend a one-ring-to-rule-them-all approach and a shift to hybrid cloud.

First things first: Defining multi-cloud versus hybrid cloud.

Before we can talk about strategy, we should define what multi-cloud is and hybrid cloud is because the terms sometimes are misunderstood in the marketplace.

Multi-cloud is really two clouds, so if you have two clouds that you work with today, then you’re in a multi-cloud environment. By the way, multi-cloud could be an on-premises infrastructure. It doesn’t necessarily need to be a private cloud or a public cloud.

Hybrid cloud is defined by bridging those two clouds together, so that they interact together. Important characteristics of hybrid cloud integration are things like high speed private connectivity.

Why hybrid cloud facilitates future workload portability

Another really big component, and you hear a lot about this if you read some of the various technology publications, is preventing cloud lock-in. That’s a really important concept because multi-cloud and hybrid cloud really embrace the concept, the future concept of what we call workload portability. In the future, we will be able to move workloads from one cloud to the other cloud in a very transparent manner. Developing a hybrid strategy that will facilitate that in the future as the technology evolves is really, really important, and we have to begin that process now.

Keys to moving from multi-cloud to hybrid cloud

First, you have to decide the right place to run the workload. Not all clouds are created equal, and being selective at where you actually run those workloads is very important. Things like performance, things like disaster recovery, geo disaster recovery are all very important in considering where our workload should live.

Next, you want to start ‘top of stack’ when you start looking at your architecture versus working from the bottom up. All too often, people begin with thinking through the infrastructure and how all those components should work. You really need to start from the top with the application or the solution.

In today’s environment, we hear a lot about cyber security. And so a good emphasis when you start talking about application down is to start with cyber security or application security requirements, and then work down to all of the various components until you hit the infrastructure.

Finally is how you manage this new hybrid cloud environment. It’s obviously not very efficient to have multiple management stacks, managing workloads in different clouds, so through this hybrid component, we want to centralize on a single stack that looks and manages every one of those components including things like the application.

One ring to rule them all

Going back to the cyber security, application performance, infrastructure performance and focusing in on a single architecture that deploys a single management stack, one that we’d say rules them all or manages them all is really, really important.

Pete Schmied headshotPete Schmied is the Senior Manager, Operations at DXC Concerto.

Speak Your Mind


This site uses Akismet to reduce spam. Learn how your comment data is processed.