Best practices for setting up replication

• Create the Primary data volumes with drive letters. Plan the size and layout of the data volumes based on the requirement of your application. You must configure the Primary and Secondary data volumes with the same name.
• If you create the Replicator Log volume manually, then make sure that you do not assign a drive letter to the Replicator Log volume. InfoScale recommends that you create the Replicator Log volume as a mirrored volume. For better performance, Replicator Log should be a mirrored-striped volume.
• To improve write performance configure the data volumes and the Replicator Log volume on different physical disks.
• InfoScale recommends that you create a Replicator Log volume of the same size on the Primary and Secondary. Size the Replicator Log appropriately to avoid overflow.
• As Volume Replicator uses Replicator Log volumes, InfoScale recommends that you do not format these volumes with any file system. When a volume is assigned for use as a Replicator Log volume, the existing file system and data is lost.
• Plan the bandwidth of the network to be used, based on your requirement.
• If you plan to replicate data over cloud, ensure that you satisfy the following network requirements to establish connectivity between the on-premises and cloud networks, or in between the same or different cloud regions:
  • A non-overlapping IP address space for network subnets is available on the primary and the secondary datacenters.
  • A virtual network (VNet) is set up in Microsoft Azure network.
  • The Volume Replicator DHCP_IP_Support tunable is set to True. This tunable determines whether or not DHCP IPs can be used while configuring Volume Replicator.

• To set up replication from an on-premises datacenter to a cloud datacenter, a virtual private network (VPN) gateway and a local network gateway is set up in the cloud network. This enables to set up a communication tunnel between the machines that are located in the on-premises datacenter to the machines that are located in the cloud datacenter.

• To set up replication between any two regions on cloud, a virtual private network (VPN) gateway is set up over the virtual network, and a connection is established between the VPN gateways for the virtual machines to communicate with each other.
• To set up replication within a same region on cloud, a VNet Peering is set up between the two virtual networks in which the machines are configured.

• You can choose to use either the UDP, TCP, or STORAGE protocol for network communication between the Primary and Secondary during replication. You can use the STORAGE protocol only for a Bunker Secondary if the storage on the Bunker Secondary is visible from the Primary host.
• Make sure that the Secondary volumes are not formatted and no other application uses these volumes. If you use these volumes for replication, all the original data on these volumes is lost.
• Avoid using RDS and RVG names with special characters with the exception of the hyphen (-) or an underscore (). The names can have a maximum of 31 characters and can begin with a hyphen (-) or underscore () character.
• If you want to replicate encapsulated volumes, ensure that the volume names and sizes on the Secondary are the same as the corresponding volume names and sizes on the Primary. The volume name is auto generated while encapsulating the volume.
• Enable the ports and services that are used for inbound and outbound communication.

Related information