What is a BDR?
We recommend having a dedicated backup management computer (a BDR) when running DR backups. This computer not only manages the image chain (via ImageManager), but also carries the burden of performing the offsite backup as well as virtualizing servers in the event of a disaster.
While we don't have any specific hardware recommendations, we do have some general guidelines on what to keep in mind while picking hardware for BDRs.
Guidelines and Considerations
Regarding virtualizing servers in the event of a disaster, at a minimum, you'll want at least 70%-80% of the resources (CPU & RAM) of the largest system you plan on virtualizing. This can be scaled up depending on the maximum number of systems the BDR could be handling at once. If you have multiple servers to be spun up, you'll want to add about a 15-25% additional for overhead.
For example, if you had 3 servers with 4 cores and 8GB of RAM each you'd want a BDR with 16 cores and 32GB of RAM. Additionally, if there are multiple servers to be spun up, 10k striped drives or SSDs would be recommended to avoid bottlenecks.
What about storage?
In addition to those CPU and RAM recommendations, we suggest having ~3.5x the amount of storage used in production for storing the local image chains. For example, if you have a system with a 500GB C volume and a 1TB D volume (total size, not space used), we'd suggest having ~5TB of space to store the image chains for that system.
If you have any questions regarding these guidelines, please reach out at firstname.lastname@example.org!