In January, I spoke at The CUBE’s Supercloud 2 event, which focused on the challenges and opportunities involved with cloud adoption, as well as what’s on the horizon for the next era of digital innovation. I sat down with The CUBE’s Dave Vellante and John Furrier, co-founders and co-CEOs of SiliconANGLE Media, for a conversation about the typical journey organizations embark upon as they adopt and integrate cloud technology into their infrastructure, applications, and services delivery.
While you can watch a free recap of the whole session here, I wanted to share a bit of the discussion about the journey I’m referring to — from “cloud first” to “cloud smart” — in today’s Whiteboard post. My goal is to explain (and hopefully, help you avoid) some of the typical missteps that I see businesses make as they begin to leverage the true value of a multi-cloud architecture.
In the diagram below, you’ll see an outline of the typical journey. It begins with “cloud first,” where the organization begins to rally around moving infrastructure and applications to the cloud. This usually takes the form of selecting a single cloud that the organization intends to standardize upon. They begin to build out functionality — utilizing cloud-specific offerings for things like disaster recovery, security, and monitoring/governance. That’s all well and good.
When these organizations start to understand the benefits of the cloud, they usually arrive at a “eureka” moment where they decide to expand their cloud implementation to other apps and services. As they do this, they begin to realize that cloud providers have different strengths. One might be best suited for certain databases, while another might optimize AI/ML services. They begin to identify best-of-breed tech and cherry-pick the offerings that best meet these needs. This is a common path to using multiple clouds. Another way we see companies arrive at multi-cloud is through mergers and acquisitions. So even the most intentional of IT organizations may inadvertently find themselves working across clouds.
This is where the “cloud chaos” phase begins.
As I’ve illustrated in the diagram below, what typically happens is that the organizations end up replicating and applying the technology stacks they’ve built to various clouds, on-premises datacenters, and edge, usually in a siloed fashion. Before they know it, they end up with multiple teams and solutions that all serve the same purpose. It’s not scalable or sustainable. And, yes — it’s chaotic.
Eventually, they start to look for solutions they can apply across all of their various clouds (public, private, edge, and so on). As they begin to centralize these operations, they move into a more mature phase of their cloud journey, employing cross-cloud services that work across the board. This is what I refer to as “cloud smart.”
The good news is that with a bit of knowledge and awareness, organizations can skip over the chaos phase and go straight from “cloud first” to “cloud smart,” more quickly realizing the benefits they envisioned when moving to the cloud in the first place.
The folks at The CUBE call this the “Supercloud,” which I’ve illustrated below (see my first Supercloud blog from The CUBE’s event last August). This is the “nirvana state,” in which the organization builds and maintains a single set of services — for applications, infrastructure, security, end-user services, and data-plane services — and implements them across all the various clouds they leverage.
The Supercloud is manageable. It’s integrated. It’s scalable. And it’s where you want to be.
Not to toot our own horn, but VMware saw this coming years ago. It’s why we built our Cross-Cloud Services — as a means of simplifying the complexity that arises as organizations move to a multi-cloud architecture. Now we, of course, recognize that VMware can’t do everything. That means that the “Supercloud” will be composed of offerings from many different vendors, creating the need for an overall architecture.
There’s more to Supercloud than just cross-cloud services, such as thinking through the difference between platforms and architecture, and I’ll share more about that in my next Whiteboard post. I’ll also talk about some of the research and emerging technologies that will influence both.
Until then, feel free to share your own thoughts and experiences regarding the journey from “cloud-first” to “cloud-smart” in the comments.