Moving Servers to the Cloud? Write the Rollback Plan Before the Migration Plan
Many cloud moves go wrong for a plain reason: the team plans how to move forward but not how to step back. If a migrated system misbehaves on day one, someone needs to know exactly how to return to the old setup without losing data.
A sensible migration usually runs in four stages, and rollback belongs in the second one, not as an afterthought.
The four stages, in plain terms
- Assess: list what you run today, what depends on what, and which systems cannot stop.
- Plan: decide the order of moves and write down the rollback steps for each one.
- Move in stages: shift a few systems at a time so problems stay small and data integrity can be checked.
- Optimise and manage: after the move, tune cost and performance and keep monitoring.
Questions to settle before the first server moves
- Should each workload sit in private, public or hybrid cloud, given your compliance, performance and cost needs?
- Is backup and disaster recovery planned, built and actually tested, not just switched on?
- Who hardens and monitors security in the new environment from day one?
- Who looks after servers after go-live: updates, monitoring and tuning?
Getting these answers on paper first keeps operations running while the move happens.
See how a staged approach works on Affix Center's page on cloud migration and infrastructure management in Mumbai.
Published by Affix Center, a Mumbai-based IT services company.
Comments
Post a Comment