Replication job settings to not use so much bandwidth

[RedShamrok]

Hello,

I am new to Veeam and backups, and am still learning how to use it, just gonna put that out there. My company has setup our DR solution to an offsite data center that we are using wan accelerators to transfer data. Currently we are facing an issue where the data center is charging us for bursting and using too much bandwidth. Not sure if it's a setting on the jobs or not that could be causing the issue.

Currently here are the settings we are using:

Wan Accelerators: 2 accelerators, one at source location one at DR site, both with same settings

• Server:
  • Streams: 5

• Cache:
  • Folder Capacity: 5TB
  • Cache Size: 100GB

Replication Job settings:

Job Settings: Each job has the same settings. (only showing the ones that are enabled)

• Traffic:
  • Exclude swap file blocks
  • Exclude deleted file blocks
  • Compression level: optimal
  • Storage optimization: WAN Target

• vSphere:
  • Use Change Block tracking data
  • Enable CBT for all protected VMs automatically
  • Reset CBT on each Active Full backup automatically

• Integration:
  • Enable backup from storage snapshots

• Data Transfer:
  • Source proxy: source
  • Target proxy: target
  • Through built in WAN accelerators

I am not sure if there is a better setting for the compression level or anything that I need to set to lower the bandwidth usage? Any input is appreciated, and I hope I gave enough information.

[wishr]

Hi Timothy,

Please, take a look at Network traffic management and let us know if that is what you are looking for.

Thanks

[foggy]

Hi Timothy, what is the uplink speed, the reported transfer speed, and the full bottleneck stats for the jobs?

[RedShamrok]

Hello, sorry for the late replies I do believe the Network traffic management will help in some aspects of this.

As for the uplink speed, I am unsure of, as it is from the data center itself, and I will have to contact them to get that info. The reported transfer speed on the replication jobs are all under 1 MB/s, and the bottleneck on most of the jobs are the Target WAN.

[foggy]

Just to make sure I understand the issue - you've initially stated that the jobs are using too much bandwidth, is 1 MB/s too much? Or are you talking about the overall bandwidth for all the jobs then? Or do you mean the amount of transferred data?