It took a village to get persistent storage into Kubernetes.

As the container orchestration tool grew in popularity, users wanted persistence or an ability to retain data after a device had been shut down. In the 2015 timeframe, Red Hat Inc. engineers convened a group of leading storage providers to find a way to add persistence while running stateful applications using Kubernetes.

“Red Hat was extraordinarily interested in it because at the time we were doing battle with Cloud Foundry,” said Mike Barrett (pictured, right), vice president of product management at Red Hat. “We brought all of these storage vendors to the table, and they all had their own idea of what that implementation should look like. You really need an ecosystem like the one we’re sitting in today, the power of Kubernetes, the power of KubeCon, the power of all these people coming together with an open heart and really driving the best outcome. That’s how we got storage into Kubernetes.”

Barrett spoke with theCUBE industry analysts John Furrier and Savannah Peterson at KubeCon + CloudNativeCon NA, during an exclusive broadcast on theCUBE, SiliconANGLE Media’s livestreaming studio. He was joined by Brandon Mann (left), product management leader for Storage Fusion at IBM Corp., and they discussed multicluster management and how AI is driving storage use cases. (* Disclosure below.)

Enabling hosted control planes

The push for persistent storage in Kubernetes was driven by growing enterprise interest in applications and an ability to manage multiple clusters. This led to further developments in areas such as hosted control planes, a key area of focus for Red Hat and IBM.

“It’s having infrastructure that can be rolled in, set up and away you go with your cluster or clusters in a matter of days, not months,” Mann said. “That’s one of the areas that we’ve really focused on at IBM. There are some key technologies that we’re developing with Red Hat and the broader Kubernetes community in terms of multicluster, and some of the things like hosted control plane that I think are fantastic enablers to really drive development and innovation.”

Growing interest in generative AI is having an impact on the storage industry as well. This is becoming more evident as vendors such as Red Hat work with companies that must rely on large and often private datasets.

“To really hit a home run with generative AI, you have to have your most intimate of data. You have to have stuff you don’t share with people to really train those models,” Barrett said. “You’re not necessarily going to put that up in the cloud today, and it’s driving a lot of storage use cases for us.”

While neither executive spoke in detail about upcoming plans for storage enhancements, Mann offered general insight into how the two companies viewed key drivers in the tech landscape, including AI.

“We’re looking to have some pieces around further integration around appliances,” Mann said. “When it comes to AI, I think there’s a lot that’s progressing between what we have with OpenShift AI and what we have with watsonx. Those two are hand-in-hand.”

Here’s the complete video interview, part of SiliconANGLE’s and theCUBE’s coverage of KubeCon + CloudNativeCon NA:

(* Disclosure: IBM Corp. and Red Hat Inc. sponsored this segment of theCUBE. Neither IBM, Red Hat nor other sponsors have editorial control over content on theCUBE or SiliconANGLE.)

Photo: SiliconANGLE