Cloud Technologies

Why Supercloud Architectures Could Upend Cloud Computing – Or Not

The Future of Cloud Computing With Instant Cross-Cloud Migration
Pinterest LinkedIn Tumblr

What Is Supercloud?

Supercloud is an approach to cloud computing that abstracts underlying cloud platforms from applications so completely that it allows applications to move seamlessly between clouds – or even operate across multiple clouds at the same time.

Thus, if you were to adopt a supercloud strategy, you’d build a cloud architecture that lets you migrate an application instantly from, say, AWS to Azure, without having to reconfigure the application or its environment in any way. You’d also be able to do things like host some of the application’s microservices on Azure and others on Google Cloud Platform (GCP) at the same exact time.

Supercloud could prove to bring massive disruption to the cloud computing industry because it opens up a host of opportunities that aren’t viable under traditional multicloud architectures.

Supercloud Versus Multicloud

To explain why supercloud could turn out to be such a big deal, let’s first talk about how it’s different from traditional multicloud.

As of 2024, multicloud architectures – which mean using multiple clouds at the same time – are commonplace. IDC’s March 2024 Cloud Pulse Survey (n = 1,350) shows that 74% of cloud buyers have multicloud strategies. It’s no longer a big deal to use multiple clouds.

However, traditional multicloud architectures simply involve using one cloud platform to host some workloads and other clouds for other workloads. They don’t deeply integrate cloud platforms together. As a result, with traditional multicloud, migrating an app from one cloud platform to another is typically a complicated process because you have to reconfigure the application to run in the new cloud. This entails tasks like rewriting identity and access management (IAM) rules, reconfiguring networking, and selecting and setting up new compute and storage services.

Likewise, the idea of hosting applications across clouds at the same time is virtually unheard of, even for organizations that have long used multiple clouds. It’s very rare to try to have an application frontend run in one cloud while its back-end components are hosted on a different cloud, for example. Network latency issues would present a big challenge if you tried to do this. You’d also need to implement application logic that allows your internal application services to connect across clouds, which would significantly complicate the application development and management process.

But supercloud could change all of this. By making underlying cloud platforms irrelevant from an application’s perspective, supercloud has the potential to take multicloud to a whole new level.

Benefits of Supercloud

Specifically, supercloud architectures could deliver benefits like the following:

  • Maximizing application reliability by hosting complete instances of an application on multiple clouds at once. This would mean that even if an entire cloud crashed, the app would keep running.
  • Optimizing cloud costs by making it possible to migrate to a different cloud instantly if better pricing becomes available in that cloud.
  • Eliminating the need for teams to learn the intricacies of multiple cloud platforms. With supercloud, cloud service vendors’ tooling and configurations would become less important because they’d be abstracted from IT operations.
  • Improving application performance by making it easy to distribute application instances across cloud platforms and regions. This would reduce latency and speed application responsiveness, resulting in a better user experience.

How Realistic Is Supercloud?

In theory, supercloud would open amazing new doors in the realm of cloud computing. But is it actually feasible in practice to build a supercloud architecture?

The answer remains unclear. Although the supercloud concept has generated a bit of chatter over the last year or two, no vendor has come close to developing solutions for actually creating a supercloud.

There are, of course, plenty of cloud monitoring, management and security tools that support multiple cloud platforms. To an extent, they smooth the process of operating applications across clouds. But they certainly don’t erase the barriers to instant cloud migration or cross-cloud operation. Being able to use the same tool to monitor applications that run in different clouds is quite different from having apps that work exactly the same no matter which cloud hosts them.

There are also some application hosting platforms that abstract applications from underlying infrastructure in ways that could, in theory, help to build superclouds. Kubernetes, the open source orchestration platform, is a prime example. Theoretically, you could build a Kubernetes cluster in which some nodes are virtual services running in one cloud, while other nodes are servers hosted in a different cloud.

But this is not what Kubernetes was designed for, and multicloud Kubernetes clusters are very rare. Building them requires grappling with complex technical issues, like the difficulty of keeping the various parts of a Kubernetes cluster in sync when they are distributed across multiple clouds and rely on the internet, instead of superfast local networks, to communicate.

So, while we do have some solutions that gesture toward a supercloud future, building a supercloud today would be a very fraught and clunky experience, at best.

What It Will Take to Make Supercloud a Reality

But the hurdles to supercloud don’t seem impossible to overcome. If cloud service providers were to collaborate around developing shared standards for configuring and using cloud infrastructure, building a supercloud would become quite easy. Imagine, for instance, that instead of having to write different IAM and networking rules for each cloud you use, or select different types of cloud server instances, you could write rules or select infrastructure that worked on any cloud. Technically speaking, this wouldn’t be too hard to do, if cloud providers got on board.

The challenge, of course, is that cloud providers currently have little incentive to make it easier for customers to use competitors’ platforms at the same time. Amazon doesn’t stand to gain anything by making it easy for its customers to migrate AWS-based apps instantly to Azure or GCP, for example.

Another possibility is for a single vendor to build a supercloud platform that abstracts underlying clouds from applications. A third-party solution could translate between different cloud service providers’ tooling and services in ways that enable a consistent application deployment experience, while also solving for the cross-cloud connectivity issues that abstraction platforms like Kubernetes don’t currently address.

But the problem there is that customers would end up locked into a supercloud platform owned by one vendor. They’d also presumably end up paying more because the vendor would effectively be reselling public cloud services and adding a premium.

The bottom line: Bringing supercloud to fruition will require solving a business challenge more than a technical challenge. The technology is feasible enough to build. Getting vendors to cooperate with one another sufficiently to enable a supercloud future is the hard part.

Christopher Tozzi, an adjunct research advisor for IDC, is senior lecturer in IT and Society at Rensselaer Polytechnic Institute. He is also the author of thousands of blog posts and articles for a variety of technology media sites, as well as a number of scholarly publications.