Renewing virtual infrastructure with recovery in mind.

Keep the platform supportable through its lifecycle

The compute platform was due for replacement in line with the manufacturer’s product lifecycle. I had to account for existing workloads and recovery requirements in the replacement specification.

With specialist supplier input, I decided server specifications, licensing, storage and memory capacity. I sized for disaster-recovery failover as well as normal operation and specified redundant network connectivity.

Design around storage and recovery dependencies

The replacement hosts needed to work with SAN storage, networking and the backup services supporting the virtual machines. I considered compatibility and the dependencies involved in migrating or recovering workloads.

My backup and recovery work spans replication, backup copies and tape-based retention. I accounted for those processes when specifying the replacement platform.

Install the hosts and move the workloads

I physically installed and cabled the replacement equipment, built the virtualisation hosts and configured storage. I then migrated live virtual machines onto the refreshed platform.

I checked the migrated services and worked with specialist suppliers where additional input was needed.

Support the platform beyond the refresh

I continue to administer the virtualisation platform, investigate storage and backup faults, maintain documentation and coordinate specialist support. Fault investigation involves tracing dependencies between the guest operating system, host, storage and network.