Off-site replication: Calculating bandwidth?

[cyberswb]

Is there a method for calculating bandwidth required to do offsite replication? The post-replication email/reports don't really provide a guideline as to the "incremental" size of the replication.

We currently do it on site to a seperate SAN located in another part of the building, but management has been told by key customers that they consider this cost saving "on site DR facility" more about cost savings and less about DR, so they are looking to overhaul the DR process and move replication to an off-site facility.

One of the initial questions that's been raised is how much bandwidth is needed for our normal replication cycle, which is 6 times a day (every 4 hours).

I'm guessing the rough way to do this is to take the "worst case" VRB file size for each of our replicated VMs, add them up, and then use that as the data amount in calculating bandwidth. Is that at least back of the envelope accurate? Is there some overhead multiplier I should add in?

[Bunce]

General rule of thumb we've found is to go by the VRB size changes and then multiply by 2 for overhead.

(see some of my previous posts for more info if required).

We're hoping for more useful statistical reporting in V5..

[Vitaliy S.]

Hello Shawn,

Bunce is correct, actual value for bandwidth needed depends on the total amoung of changes for VMs being replicated. You may consider looking through this thread, as it addresses the same question:
http://www.veeam.com/forums/viewtopic.p ... ics#p13420

Thank you!