Infrastructure strategy is tilting toward unified application intelligence as enterprises try to keep up with increasingly mixed workloads across data centers and public clouds.

Hybrid patterns are now the norm rather than the exception, and organizations are looking for ways to run virtual machines, containers and serverless side by side without multiplying complexity, according to Sudeep Goswami (pictured, right), chief executive officer of Traefik Labs Inc.

Traefik’s Sudeep Goswami and Nutanix’s Dan Ciruli talk with theCUBE about how new infrastructure layers and partnerships aim to make hybrid and multicloud portability a practical reality.

“The learning period is over, the experimentation is over,” he told theCUBE. “Now, people are thinking seriously about optimizing workloads for different tasks, and this coexistence of [virtual machines] and containers are here. But then we’re also seeing serverless come in. You have to think about — and this is on-prem we’re talking about, not just cloud, but everywhere. You got to think about these co-existences. This requires … [taking] a step back, and you think about what is the right architectural decisions that you have to make to enable this at velocity now.”

Goswami and Dan Ciruli (left), general manager of cloud native at Nutanix Inc., spoke with theCUBE’s Rob Strechay during the KubeCon + CloudNativeCon NA event, during an exclusive broadcast on theCUBE, SiliconANGLE Media’s livestreaming studio. They discussed how new infrastructure layers and partnerships aim to make hybrid and multicloud portability a practical reality for demanding enterprise workloads. (* Disclosure below.)

Unified application intelligence drives new hybrid infrastructure choices

Traefik’s newly launched Application Intelligence Platform serves as a higher-level control plane that unifies VMs, containers and serverless under an application-centric decision layer, according to Goswami. Running alongside Nutanix’s Kubernetes-based infrastructure foundation, it brings serverless-style experiences on-prem without creating new silos. The unified application intelligence model abstracts away fragmentation across environments while maintaining policy and governance consistency.

“We have done a great integration with Nutanix,” Goswami said. “The way we look at this is the way to solve the fragmentation is through abstraction, and you create the right abstraction layer. As the traffic is coming in to go to different places, the application layer logic decides whether that’s going to a VM, whether that’s going to a container or to a serverless, and you manage all of it through a consistent policy layer.”

From Ciruli’s vantage point, the pain points emerge when enterprises run VMs in one environment, containers in another and Kubernetes on bare metal, each with different networking and security workflows. That siloed pattern complicates traffic routing, role-based access control and operations at scale, prompting teams to look for a way to run virtualized and containerized workloads, along with storage, on a single stack rather than stitching together multiple infrastructure islands.

“One of the silos that we see happening frequently is that people are … running their VMs in one place, made literally some set of clusters that are virtualized and running their containers in another place, running Kubernetes on bare metal,” Ciruli said. “It leads to big issues: How you direct the traffic, how you do the networking between them, how you do the RBAC between them? Because you have entirely different workflows, entirely different ways to handle all of that.”

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

(* Disclosure: TheCUBE is a paid media partner for the KubeCon + CloudNativeCon NA event. Neither Red Hat Inc., the primary sponsor of theCUBE’s event coverage, nor other sponsors have editorial control over content on theCUBE or SiliconANGLE.)

Photo: SiliconANGLE