Red Hat launched the latest iteration of its long-running OpenStack platform with new telecom operator-focused features that further closes the bond between that product and the vendor’s cloud-native OpenShift platform.

Specific OpenStack 17 updates include dynamic resource allocation and director designed to support “lighter” deployments. These are applications that don’t require as many services and are targeted at a smaller infrastructure footprint like edge or resource-constrained locations.

There is also an open virtual network (OVN) migration tool to validate migrations between open vSwitch (OVS) and OVN; hardware offload support for smartNICs; and a tech preview of a role-based access control (RBAC) with increased enforcement across OpenStack services.

The update also tightens integration between OpenStack and OpenShift. This includes the ability to run both virtual network functions (VNFs) and cloud-native network functions (CNFs) side-by-side in the network.

Furthering that integration, OpenStack 17 supports a virtualized control plane that can ride on top of OpenShift. Maria Bracho, who leads Red Hat’s OpenStack team, explained this allows users to run cloud native and virtual machines (VMs) within the same infrastructure.

“So now OpenStack control plane only is the workload on top of OpenShift,” Bracho said. “And the reason for that is because we have customers that want to run OpenShift on bare metal for some workloads but want to leverage the orchestration of OpenShift to also manage the control plane of OpenStack.”

Bracho noted this work further tightens the vendor’s OpenStack and OpenShift platforms, adding that customer conversations are increasingly tied to the use of both OpenStack and OpenShift depending on workload requirements.

“We have found in our later releases over the last couple of years that the conversation is not a religious dispute between the two platforms but rather where their workloads are served best,” she explained. “We have done a ton of work to make sure that the platforms can be used together, that commercially it's also a non-issue which platform you choose for the workload that you need, and in particular, for deployment models.”

This tightening is important for telecom operators that want to tap into OpenShift’s quicker innovation pace but need longer-term support afforded by OpenStack.

“OpenShift has a shorter lifecycle,” Bracho said. “Workloads are expected to move quicker, to be more resilient, to be certified, to be set up quicker, and that's not something that [virtual network functions] are doing today. All this conversation on OpenShift, OpenStack, it's all about the workloads.”

A recent ABI Research report grouped Red Hat and VMware as the top 5G telco cloud-native platform providers.

“Their goals in telco are similar, and they are both positioning themselves to be multi-vendor, cloud native, and multicloud with the capability to service the core, edge, and RAN,” ABI Research analyst Kangrui Ling explained.

Red Hat’s OpenShift was specifically touted for its auto-scaling of clusters and zero-touch provisioning through the vendor’s Red Hat Advanced Cluster Management for Kubernetes, Red Hat Ansible Automation Platform, and Operator Lifecycle Manager (OLM).

Despite the OpenShift leanings, OpenStack remains a stalwart of current 4G LTE deployments, thus the need for ongoing updates. Red Hat claims OpenStack, which has been on the market for more than 10 years, is used by more than 30% of commercial 4G LTE networks, including ongoing work with Verizon using both OpenStack and OpenShift.

“While there's a lot of talk about 5G, not every country or provider is ready for 5G,” Bracho said. “There's a ton of 4G being deployed even on large carriers. And in the U.S., for example, there's a ton of 4G still being deployed. So there's a ton of OpenStack planned and being deployed today and tomorrow.”

Catch up on all of our coverage from MWC Las Vegas here.