This article is based on reporting originally published by nextplatform.com, drawing on an interview with Brian Gracely, then Senior Director of Product Strategy for Red Hat’s OpenShift technology.
The Hybrid Cloud Reality
Despite more than a decade of rapid growth by hyperscale public cloud providers — Amazon Web Services, Microsoft Azure, and Google Cloud — the majority of enterprise workloads still run in on-premises datacenters. Security requirements, compliance mandates, data gravity, and sheer organizational inertia mean that legacy infrastructure is not disappearing any time soon.
According to Brian Gracely, speaking to The Next Platform, roughly 80 percent of enterprise workloads and 80 percent of enterprise IT spending remained behind the firewall at the time of publication, with the remaining 20 percent running in the public cloud. That breakdown is shifting, but gradually — and it is why hybrid cloud and multicloud strategies have moved from aspiration to operational reality for most large organizations.
The trend line is clear: public cloud datacenters now account for more than a third of the total datacenter hardware and software market, and enterprises are becoming more comfortable moving even mission-critical workloads — ERP systems, SAP environments, financial applications — into the cloud. Meanwhile, offerings like AWS Outposts, Microsoft Azure Stack, and Google Anthos reflect public cloud providers extending their platforms into enterprise datacenters, blurring the on-premises/cloud boundary further.
Red Hat’s Hybrid Cloud Bet
Red Hat, acquired by IBM in 2019 for approximately $34 billion, has built its modern product strategy around the open hybrid cloud model. The core idea is straightforward: enterprises need consistent tooling — for security, developer workflows, operational management — regardless of where their workloads actually run. That consistency is the problem OpenShift is designed to solve.
“Everything that we announce — at least that’s not just about Red Hat Linux — tends to be in that open hybrid cloud framework. If you look at the high-level things when IBM acquired us, that was really their mantra as well. They believe the world is in early stages of overall cloud adoption. They believe in hybrid cloud and they want Red Hat to help them drive that vision.”
Brian Gracely, Senior Director of Product Strategy, Red Hat OpenShift
OpenShift: Enterprise Kubernetes at Scale
OpenShift is Red Hat’s enterprise-grade container and Kubernetes platform. Originally released approximately five years before this article’s publication date (around 2015), OpenShift ships on a release cadence of roughly every three to four months. It builds on upstream Kubernetes — the container orchestration system originally developed inside Google and later donated to the Cloud Native Computing Foundation (CNCF) — and layers on enterprise features including integrated CI/CD tooling, role-based access control, multi-tenancy, and operator-driven lifecycle management.
At the time of publication, Red Hat reported that its OpenShift customer base had grown from approximately 1,000 to more than 1,700 customers, with strong adoption among Fortune 2000 companies. Notable examples cited by Gracely included:
- BMW — using OpenShift for workloads related to autonomous vehicle development and simulation.
- Ford — using OpenShift to enable internal IT teams to deliver AI and analytics capabilities as a service.
OpenShift 4.4: What Changed
Announced at Red Hat Summit 2020 (held virtually due to the COVID-19 pandemic), OpenShift 4.4 was built on Kubernetes 1.17. The release focused on three areas relevant to hybrid cloud operations:
- Improved developer observability — better built-in tooling for monitoring platform health and application workload performance.
- Cost and resource management — visibility into resource consumption across hybrid cloud environments, helping teams track and optimize spending at the application level.
- Edge readiness — continued investment in supporting deployments that stretch from traditional datacenters to public cloud regions and out to edge locations.
Advanced Cluster Management for Kubernetes
One of the more operationally significant announcements at the same Summit was Red Hat Advanced Cluster Management (ACM) for Kubernetes, released initially as a Technology Preview. ACM addresses a real and growing pain point: enterprises running Kubernetes at scale often operate dozens — sometimes hundreds — of separate clusters spread across on-premises infrastructure and multiple public clouds. Managing these clusters individually does not scale.
ACM provides a centralized control plane for cluster lifecycle management. Key capabilities include:
- Multi-cluster visibility — a single pane of glass for the state of all clusters, regardless of where they run.
- Policy-based governance — define security and compliance policies once and propagate them across all clusters. For example, a financial services organization can apply regulatory policies to every cluster in its fleet from a single configuration.
- Application lifecycle management — deploy and update applications across multiple clusters without per-cluster manual intervention.
- Scale — designed to manage thousands of clusters spanning on-premises datacenters and public cloud providers.
“That gives us a really scalable management technology to manage thousands of clusters no matter where they run. It reduces a ton of complexity for customers. It allows them to do kind of central policy management across clusters. So if you want to apply your financial services regulatory policy, you can deploy that across all of them.”
Brian Gracely, Red Hat
Unified VM and Container Management via KubeVirt
Perhaps the most architecturally interesting announcement was OpenShift’s expanded support for running virtual machines (VMs) alongside containers and serverless workloads — all within a single Kubernetes-based platform. This capability, available as a Technology Preview at the time, is built on the KubeVirt open-source project, which allows VM workloads to be defined, scheduled, and managed using standard Kubernetes APIs.
The practical implications matter for enterprise operations:
- Fewer management platforms — organizations currently maintaining separate platforms for VMs (VMware vSphere, for example) and containers can consolidate onto a single OpenShift environment.
- Reduced licensing costs — replacing proprietary hypervisor licensing with open-source virtualization can reduce infrastructure costs for eligible workloads.
- Developer self-service for VMs — historically, requesting a VM meant opening a ticket with an infrastructure team and waiting days or weeks. With KubeVirt integration in OpenShift, developers can request VMs through the same self-service workflows they use for containers, dramatically reducing provisioning friction.
- Support for legacy software — not everything can be containerized. Some packaged enterprise software is only supported by vendors when running inside a VM. Unified management means those workloads don’t require a separate operational silo.
“What this does is brings virtual machines into a completely self-service mode for developers. The same way that they said, ‘Give me access to containers and Kubernetes resources,’ if part of that application requires something that’s a virtual machine, they now just do a self-service request. It has the potential to take a lot of friction for developers that have a mix of new parts of an application along with some existing parts of an application.”
Brian Gracely, Red Hat
Key Takeaways for DevOps and Platform Teams
The direction Red Hat is taking with OpenShift reflects broader industry realities that platform engineers and DevOps practitioners need to account for:
- Hybrid cloud is the operational default, not a transition state. Design for workloads running in multiple locations from day one.
- Kubernetes cluster sprawl is a real management burden. Multi-cluster management tooling (ACM or equivalent) should be part of any serious Kubernetes platform strategy.
- VMs are not going away. Platforms that can manage both containers and VMs coherently will reduce operational complexity more than platforms that force an all-or-nothing migration.
- Policy-as-code at scale is becoming a first-class requirement, particularly in regulated industries. Centralized governance that propagates to distributed clusters is essential for compliance.
- Developer experience matters at the infrastructure level. Reducing VM provisioning from weeks to minutes through self-service has real productivity impact.
Note: This article reflects the state of OpenShift and the broader hybrid cloud market as of April 2020. Specific product features, customer counts, and market figures may have changed substantially since publication. Consult the Red Hat OpenShift product page for current capabilities.