FlashSystem HA and DR Architecture
How to frame IBM FlashSystem high availability and disaster recovery questions before choosing replication, site layout, and recovery tooling.
High availability and disaster recovery are different promises. A FlashSystem design can reduce storage outages without solving application recovery, and a DR design can replicate data without proving that IBM i, VMware, or database services recover cleanly.
RPO
How much data can the business lose, by workload and by time of day?
RTO
How long can each workload be unavailable while storage, hosts, applications, and users fail over?
Site model
Decide local HA, metro distance, remote DR, or vault isolation before choosing replication methods.
Test method
A DR plan is not real until a restore or failover has been tested without damaging production.
For IBM i environments, include save strategy, journaling, application dependencies, and host-level recovery in the same conversation as FlashSystem replication.