Hybrid Cloud Deployment Model Using OpenShift for Enterprise Application Modernisation

Abstract

Regulated enterprises running heterogeneous application portfolios face a genuine operational bind: workloads anchored to on-premises infrastructure by data-residency law or latency requirements cannot simply be relocated to public cloud, yet the cost and engineering overhead of maintaining parallel provisioning models is unsustainable at scale. [1,2] This paper describes a reference architecture and an implementation framework built around Red Hat OpenShift Container Platform that treats hybrid cloud not as a transitional state but as a long-term operational target. [4] The study draws on an eighteen-month longitudinal engagement with three production environments spanning financial services, healthcare informatics, and e-commerce, tracking eighteen workloads that together processed 2.4 million monthly transactions. We applied a five-stage modernization pathway—from portfolio assessment through cloud-native re-architecture—based on the strangler fig decomposition strategy and Kubernetes-native tooling, without mandating wholesale platform replacement. [7,15] Measured outcomes at month 18, benchmarked against DORA elite performance thresholds, [22] showed deployment frequency rising from 1.2 to 4.9 releases per week (+312%), change failure rate falling from 18.7% to 4.2%, mean time to recovery cut from 4h 22m to 1h 34m, and normalized infrastructure cost reduced by 38%. This paper reports those results in full, explains where the architecture produced unexpected trade-offs, and identifies the organizational conditions that most strongly predicted progression speed. [19]