A: Hyper-V Replica sends the log files or write changes for a virtual machine (VM) to the replica Hyper-V server for merging to the target virtual hard disk (VHD) every five minutes, and this interval can't be changed. If you want to have more granular control of the bandwidth used by Hyper-V Replica, you have two options.

  • Change the port used for the Hyper-V Replica from the default 80 or 443 (which can be done in the GUI or via Windows PowerShell), then use Quality of Service (QoS) to limit bandwidth at different times of the day for that port. This effectively throttles the transmission of the write logs. If insufficient bandwidth was allocated, then the logs would build up until the QoS policy allocated more bandwidth, such as at night, when the logs could be transmitted and catch up.
  • Use a script to pause replication at certain times, then resume when desired.

Both options need care, because if too much change occurs, the transaction log queue could be very large, taking a long time to send. If if the log size is more than 50 percent of the total VHD sizes, then a full resync operation will be performed. This takes additional IO and time as a block-by-block comparison is performed.