Our client, one of the big three utility firms, was undertaking a large scale physical data centre relocation. Having built two brand new purpose built data centres, the challenge was to migrate their existing server estate and infrastructure to the new locations. As these servers ran all the back office, customer facing and safety critical systems, minimising the impact on the business and mitigating any risks was of paramount importance. We were engaged to implement and run a risk mitigation process, and plan each individual server move.
We needed to convince senior stakeholders that we knew exactly what would be involved in the relocation of each server, and that all risks had been mitigated. Without signoff from the senior stakeholders, server moves would not be permitted to take place.
Prior to relocating a server, we ran a number of tests to confirm the servers and architecture operated as per the design, and that we fully understood the impact on the business.
This was an extremely time consuming task, however it was necessary to mitigate any risks associated with running on backup systems during the physical relocation of servers. The relocation exercise was completed successfully, and in the process, the entire server estate benefited from a shakedown and comprehensive failover testing.
What we did
- We worked with the existing IT infrastructure teams to implement an “infrastructure passport process”. This process required:
- Every server to be audited, detailing what each server hosted and the business owners.
- Documentation detailing the architecture of each server/hosting solution (e.g. load balanced clusters, active/passive, mirrored) and the operating capacity.
- A detailed failover/disaster recover plan for each server.
- Detailed relocation steps, detailing every step, the duration and person responsible.