Low replication speed between Netbackup domains

Good afternoon!
There is a problem. There is a low data transfer rate between the main Netbackup domain and the remote Netbackup domain.
The transfer rate is approximately 5-10% of the total channel bandwidth.
Low speed is observed when performing backup vmware, file system, and, finally, -auto image replication.
The probable cause of low speed through the data transfer channel is latency.
Perhaps colleagues have experience working with a channel that has high latencies?
What transport and application level parameters are applied, including?
What set of Multi-Threaded Agent parameters do colleagues use?
Please share your experience.
For information. Parameters for nbd transport, which are applied on media servers:
\Common Files\VERITAS\VxMS\Shared\VDDK\bin\VixDiskLib.ini
vixDiskLib.nfcAio.Session.BufSizeIn64KB = 16
vixDiskLib.nfcAio.Session.BufCount = 4