- Challenge
- Upgrade a bespoke Oracle application stack
- Assumption
- A like-for-like deployment was the safest route
- Decision
- Redesign the deployment and operating process
- Outcome
- Days of recurring effort reduced to hours, with automated teardown
A client operated a bespoke application on an Oracle platform. One supplier maintained the application while the incumbent support provider was responsible for the surrounding environment and day-to-day operation.
The support requirement included creating a complete duplicate database stack every month. Preparing it normally consumed around two days of specialist engineering effort, followed by another day to dismantle the environment after use. The activity was not a compliance requirement. It had simply become part of the accepted operating model.
When the platform needed to be upgraded, the obvious route was a like-for-like replacement. That would have delivered a newer version while preserving the same recurring cost, manual effort and specialist dependency.
We challenged that assumption and proposed a different deployment approach based on our Oracle and Linux experience. It departed from the incumbent design, so we worked through the risks and practical differences with the client before agreeing the change.
The client-facing outcome did not visibly change. The monthly duplicate environment was still provided and the application team continued to receive what it needed.
The operating model changed permanently. Environment preparation fell from roughly two days to around three hours, and teardown became automated. A repeated multi-day task became a controlled, repeatable process with far less specialist effort.
The Oracle upgrade also removed a recurring operational cost that would otherwise have continued indefinitely.
A technical change should leave the organisation easier to operate. In this case, a task that had consumed several specialist days each month became a short, repeatable process.
Engineering lessons
- A like-for-like upgrade can preserve avoidable operational cost.
- The most valuable outcome may be invisible to users but permanent for operations.
- Deep platform knowledge creates options that a standard migration plan will not reveal.
Read the engineering principles behind this work →
Confidentiality: Engineering Notes are based on real engagements. Client identities, timelines and identifying details may be changed to protect confidentiality. The engineering decisions and lessons remain representative of the work undertaken.