To close our series on optimizing BC/DR KPIs, let's look at the recovery window and system redundancy.
Both depend heavily on how many resources you can allocate to your BC/DR solution, so optimizing these KPIs essentially comes down to getting the best value for the price.
This matters even more in the cloud. Every gigabyte you allocate is multiplied by the number of downstream environments, and over time that can inflate your cloud bill significantly.
General guidelines
- Be very careful with auto-scaling technologies. Remember that some resources, once allocated, can't be taken back.
- Use incremental backups. Full backups help reduce RTO, but they can increase storage usage dramatically.
- Consider compression to save storage space.
- Keep older data that you don't need immediate access to on cheaper, archive-class storage, such as cloud offerings like Amazon S3 Glacier, or on tape.
- Don't neglect system housekeeping, including optimizing how your databases use space. Make sure the gigabytes you allocate are actually used.
- Apply the resiliency techniques from the part on RPO, such as cold standby and multiplexing, to keep the cost of resiliency down.