Client Overview

The customer is a financial institution running its Core Banking Solution (CBS) across three geographies — Dubai, London and Sydney — each with its own dedicated Data Centre and Disaster Recovery capability.

Sunfire designed and implemented a six-cluster Kubernetes architecture for CBS, integrating Kubernetes, Oracle database replication and persistent storage into one resilient system. More than two years later, the platform remains in production, has passed multiple regulatory audits, and proves its resilience every month through a live DC-to-DR switchover.

Challenges Faced

• Running core banking on Kubernetes demanded availability, data protection and governance far beyond basic container deployment.
• Dubai, London and Sydney each needed their own production and disaster-recovery capability.
• Database replication and persistent storage had to be engineered into the platform, not bolted on afterward.
• A disaster recovery design is only as credible as its last real switchover.
• Financial-services audit and governance requirements apply to every layer of the architecture.
• The platform had to prove itself over years of live operation, not a pilot.

Solutions Provided

• Understood the full requirement — geographic distribution, database replication, persistent storage, disaster recovery and regulatory governance — before designing the architecture.
• Designed six dedicated Kubernetes environments, with a Data Centre and Disaster Recovery cluster in each of the three geographies.
• Deployed CBS on VMware's Kubernetes platform, the predecessor to today's VKS, across all six environments.
• Established Oracle log-shipping replication and NFS-based persistent storage replication so DR stays continuously synchronized with DC.
• Built governance and audit-readiness into the operating model from day one, not as a later retrofit.
• Instituted monthly DC-DR switchovers that continuously validate — and improve — the recovery architecture over time.

Outcomes

• The platform has run in continuous production for more than two years without significant issues.
• DC-DR switchovers are performed every month, turning disaster recovery into a proven operational capability rather than a paper plan.
• The environment and its operating processes have successfully passed multiple regulatory audits.
• Kubernetes, Oracle replication and persistent storage now work together as one integrated, resilient architecture across three geographies.
• Audit-readiness is built into daily operations rather than assembled ahead of a review.