Through OpenShift, Red Hat Aims to Advance Hybrid Cloud Computing
Red Hat is working on a layer of software that will abstract away a lot of complexity associated with deploying the Kubernetes-based OpenShift development environment across public and private clouds.
Ashesh Badani, senior vice president for cloud platforms at Red Hat, says that each cloud platform has its own set of application programming interfaces (APIs) for surfacing compute, storage and networking resources. The layer of software that Red Hat is building will provide a unified method of deploying OpenShift that eliminates any possibility of an IT organization having to worry about finding itself locked into a cloud platform.
As part of that approach, Red Hat this week at the Red Hat Summit 2019 conference announced it will create a reference architecture for deploying OpenShift in software-defined data centers constructed using VMware vSphere, vSAN and NSX software. OpenShift already supports those VMware offerings. The reference architecture will make it easier to deploy OpenShift in those environments in part by leveraging the layer of software for abstracting cloud deployments that Red Hat is currently working on, says Badani.
While OpenShift already can be deployed on any cloud, each instance is unique in that the platform has been integrated with the software-defined compute, storage and networking services each cloud platform provider exposes via its own application programming interface. Red Hat has embarked on an ambitious effort to add a capability to OpenShift that enables the platform to automatically recognize whatever APIs are exposed by the cloud platform. That capability would make it possible for IT organizations to lift and shift entire instances of OpenShift between cloud platforms as they see fit. In fact, it’s conceivable that IT organizations will leverage OpenShift as tool to force cloud service providers to compete more aggressively for their business because the cost of moving a workload from one platform to another has been reduced sharply.
It should also make it easier for IT organizations to construct hybrid cloud computing environments around multiple instances of OpenShift. That capability should also go a long way toward fulfilling one of the promises Red Hat and IBM made when it was revealed last year that IBM plans to acquire Red Hat.
Red Hat today has more than 1,000 customers that have deployed OpenShift, which Baldani notes can be deployed on public cloud, virtual machines and bare-metal servers. As IT organizations continue to transition toward cloud-native applications running on top of OpenShift, many of them will need to be able to support OpenShift on multiple infrastructure platforms at the same time for different use cases. In fact, the expectation is that many more instances of Kubernetes will soon be deployed on bare-metal servers to reduce IT operational overhead.
It may take a while for Red Hat to realize its hybrid cloud computing vision. Right now, most IT organizations are still trying to cope with the inherent complexity associated with managing multiple clouds. But there may come a day when there is finally one set of tools and frameworks to rule them all.
