After years of effort, it appears that some progress is finally starting to be made in terms of making the OpenStack management framework accessible to the average IT organization. Red Hat today announced the availability of Red Hat OpenStack Platform 10, an instance of the management framework that is based on the latest Newton release of the open source project.
Radhesh Balakrishnan, general manager for OpenStack at Red Hat, says in addition to improved higher availability and enhanced network performance, this latest release provides a graphical interface through which IT organizations can more easily customize exactly what OpenStack components they want to deploy. That approach makes it also possible to invoke automatic upgrade capabilities to make it simpler to keep OpenStack components as current as the IT organization prefers. OpenStack compute, storage and networking services can also be customized in a way that enables IT organizations to manage them more independently of each other.
In general, Red Hat positions its distribution of OpenStack as a means for deploying Docker containers alongside an instance of the Kubernetes container orchestration engine that provides the foundation layer for the OpenShift platform-as-a-service environment. OpenStack itself is most often deployed on top of the kernel-based virtual machines, but it can also be deployed on a bare-metal server.
Balakrishnan says that Red Hat has been working with internal IT organizations, cloud service providers, carriers and managed service providers (MSP) to make OpenStack easier for each of them to deploy on their own terms. In addition, Red Hat has been working with Cisco and Dell EMC to co-engineer turnkey OpenStack platforms for the enterprise as well as platforms for carriers that are being co-engineered with Nokia, Alcatel-Lucent, Ericsson and Cisco. On the MSP front, Balakrishnan says Rackspace and Cisco have been the most active.
“It’s a holistic approach,” says Balakrishnan.
But as container technologies continue to evolve, many IT organizations are starting to evaluate the merits of running containers on bare-metal servers without employing OpenStack at all. Instead, Kubernetes provides the means for unifying the management of compute, storage and networking. It’s unclear how many OpenStack functions might be replicated one day in Kubernetes. But by and large, the open source community is trying to meld together the best of both projects to create an instance of OpenStack that can be more easily deployed using containers.
In meantime, IT organizations would be well advised to keep track of where containers wind up being deployed in 2017. Right now, most containers are deployed on top of virtual machines or in a platform-as-a-service (PaaS) environment. But as the tooling surrounding containers continues to mature, many more IT organizations are expected to opt to deploy containers on bare-metal servers to drive higher utilization rates.
That doesn’t mean that virtual machines are going away. But it does mean there will be a lot more options available for deploying application workloads in the data center.