Just as fast as the virus spread across the globe so too has demand for what Chad Dunn, VP of HCI product management at Dell Technologies, calls the Swiss army knife of the modern data center: hyperconverged infrastructure (HCI).

HCI is making its mark as a way to provide organizations with a cloud-like environment while maintaining closer control over IT resources. That pattern, Dunn said, is set to continue as vendors develop closer ties to the cloud computing model.

Dunn expects HCI technology will continue to evolve and drive major trends forward in 2021 across four areas: cloud-delivered consumption models, hybrid and multi-cloud acceleration, edge computing use cases, and cloud-native adoption.

It all boils down to the needs for speed and flexibility to add and remove resources as needed. 

Taking the titanic amount of data being generated daily into consideration, the mounting move to consumption-based, as-a-service, on-demand cloud platforms offers an avenue to greater flexibility, control, and choice over resources. These models, Dunn said, also stand to be a great equalizer, specifically for organizations that are modernizing their infrastructure at a fraction of the budgets of their larger enterprise competitors. 

“That's why HCI has a big role to play in this," he said. "It can be deployed in so many different shapes and sizes and capabilities, from the storage heavy to GPU heavy or CPU heavy, then you can use the same equipment and the same API's operating model to support that.”

Workloads are running everywhere, and IT infrastructures need the capacity and ability to flex resources that might have been available in offices, available at the edge – and that’s hybrid cloud

“We're not alone in saying this: It's going to be a multi-cloud world, and not just three or four large clouds, but many, 1000s of clouds, and that's certainly what we're already seeing,” Dunn said. “Most of the larger customers that I work with today have a hybrid cloud strategy, and in fact, usually have multiple public cloud vendors in addition to the private cloud.” 

Underneath the covers, however, that hybrid cloud model doesn’t work well without a common operating system. HCI is the seam that stitches it all together, thus providing organizations a common operating system to run on-premise hardware or in the public cloud.

To that end, he said it's not going to be uncommon to see an arrangement where “I would have a couple of public cloud providers, Dell hyperconverged infrastructure as my public cloud, and use Kubernetes to orchestrate the workloads in containers to move those services around.”

Kubernetes will be the common thread running cloud-native workloads both on-premises, as well as in public cloud deployments, Dunn said. “That's been another big driver for HCI this year: the adoption of cloud-native. 

If you’re a vendor today, and you're going to write an application from the ground up, “chances are you're going to start cloud-native right out of the gate because you want to run it in different availability zones,” he explained. Simply put: modern applications are going to be microservices oriented. 

In turn, Kubernetes will be used as a common management tool for cross-cloud scenarios. “It's going to be the lingua franca for cross cloud, for sure,” he added. “This also becomes a key driver for what’s happening at the edge.” 

Gartner forecasts that in five years, 75% of data is going to be outside of the data center versus inside. “That's basically the inversion of what it looks like today,” Dunn said. “That also means the workloads that are running out there are likely fairly new workloads and cloud-native, so we need infrastructure that can reach far edge locations and be economical, but push more compute power,” he added.

Big data and other sophisticated tasks increasingly are performed at the edge where power and space often are limited. HCI addresses the squeeze by combining computer, network, and storage with an integrated software stack that provides orchestration and virtualization of these elements.

Data is being created, moved, manipulated, and stored at astonishing levels that are only going to continue, if not accelerate, as everything from the internet of things (IoT) to autonomous cars gain more momentum.

Dunn said this will drive the need for smaller and more cost effective platforms that can operate at the edge, but also support cloud-native workloads. This signals a shift in terms of the way infrastructure is built, he added. “Infrastructure as a code base with zero touch provisioning and serviceability are going to become very important when we talk about edge deployments where there could be 10,000 plus locations for the year. You need to be able to do just about everything except physically unplug it.”