Red Hat released the latest iteration of its OpenShift platform with general availability of virtual machine management tools, Kubernetes cluster management features, and the option for a smaller operating footprint targeted at low-resource edge environments.

Red Hat’s OpenShift 4.5 platform now offers general availability of the vendor’s Virtualization feature that supports management of virtual machines (VMs) alongside containers using the same Kubernetes management layer. The feature taps into the KubeVirt open source project in allowing organizations to develop, deploy, and manage applications that use VMs, containers, and serverless from a single, container-based platform as if they were a standard Kubernetes object. This allows an organization to bring VMs into an OpenShift environment and migrate applications that are running in VMs into a container-based environment as needed.

Brian Gracely, senior director product strategy, explained that this feature allows customers to begin modernizing their VMs toward a more cloud-native model much quicker than has traditionally been possible. And it also cuts down on the over-provisioning of VM resources.

“We were hearing people say [with] our bank it takes three months to get a virtual machine. It takes three months to get through the process because we never designed it to go quickly. So what happened was developers were like alright if I know it takes that long I'm going to ask for the biggest virtual machine I could possibly get, and there was sort of a lot of inefficiency that was going on,” Gracely said. “So now we're basically [saying] this is going to be one-click away just like your containers have been. Just the prospect of taking all that friction out of getting whatever the right resources has been, it's been sort of an eye-opening thing for a lot of companies.”

Gracely differentiated this new option from what VMware is doing with its Kubernetes-focused Tanzu platform by noting a difference in direction. He explained that VMware continues to view its cloud-native push from the VM-level of infrastructure upward, while Red Hat views it from the container level of infrastructure abstraction downward.

“On the surface, yeah there is a lot of the same technology, but I think a very different way of looking at are we trying to move faster or are we trying to make things look like what we've known for a long time,” Gracely said, later adding that “when you bolt it back into an older way of doing things you haven't fixed the motivation problem, you haven't fixed the culture and the process problems.”

The virtualization feature, which takes the place of Red Hat’s previously named container-native virtualization (CNV) platform, also targets telecommunication operators that are starting to deploy 5G networks. As part of its preview launch earlier this year, Matt Hicks, SVP of engineering at Red Hat, said those operators need the ability to run OpenShift on bare metal but are also dealing with virtual network functions (VNFs) that are built on a virtualized platform.

The OpenShift 4.5 release also includes general availability of Red Hat’s Advanced Cluster Management for Kubernetes platform. That product provides a single control point for the monitoring and deployment of OpenShift clusters at scale and supports policy-driven governance and application lifecycle management for those clusters. This support runs across OpenShift on-premises, on bare metal, or on the major cloud hyperscalers.

Gracely explained that the management tool was based on a multi-cluster management product that its parent company IBM had developed. But, he noted that Red Hat plans to take the subset of that project that it’s bringing to market and will open source it at some point.

Red Hat OpenShift for the Edge

OpenShift is also now able to scale down its Kubernetes footprint to help support the increased interest in using containers and cloud-native functions in edge environments. This includes support for three-node clusters instead of the previous four-node cluster size. Gracely said the alteration involves combining the control plane and the data plane.

Red Hat aims to move OpenShift out of the data center and public cloud environments and into those edge environments.

“It’s about how do you deal with disconnected environments that are beginning to kind of spawn out of being different from big data centers, big cloud environments, and this is this is more of a footprint announcement in that context,” Gracely said.

Gracely linked the move to a conversation at the vendor’s Summit event between Red Hat CEO Paul Cormier and Srini Kalapala, VP of technology and supplier strategy at Verizon, where they discussed the need for platforms that could support a near edge deployments.

“When you see, for example, Verizon talking to us about this, we're working with them, learning and going hey we think this is the right starting point and then we sort of learned in reality how their 5G apps work, how their ops teams work with it,” Gracely said.