Replication Best Practices

[stfconsulting]

Hey Guys, we are working on getting Veeam 8 implemented in our NJ datacenter. We have about 150 servers that we need to replicated to our SC facility. We will be replicating servers directly into VMWARE Clusters down there. With that many servers how would you design the replication job(s)? Originally I was thinking about 1 per client however I am starting to see how that could become cumbersome. Curious how other people are doing this at this scale.

[Shestakov]

Hello,
I would suggest to choose several servers with similar OSes, Applications to one job and use WAN Acceleration to have better deduplication ratio.
Also compacting servers with same job settings into a single job makes sense in terms of ease of management.
This part of the user guide contains a lot of additional valuable info including Advanced Replication Technologies made to minimize infrastructure workload and reduce data traffic. Please read.
Thank you!

[stfconsulting]

Thanks for the links. Does anyone have real world #s from the WAN Acceleration? Curious what type of reductions people are seeing.

[Shestakov]

It varies depending on global cache size and how replicating data itself changes over the time. It can give up to 50x saving.
Let`s see if others have their practical results.