Planning policy-based replication

Before you can implement policy-based replication, determine your current recovery point objectives, applications requirements, and verify your networking configuration.

Before you begin

Administrators need to examine the current (and the projected future) data volumes and applications before implementing a replication solution.

  • For all applications that use policy-based replication, determine the recovery point objective (RPO) and recovery time objective (RTO). Policy-based replication supports only non-zero RPO and RTO.
  • For policy-based replication, determine the number of recovery copies that are needed and the geographic distance between the production and recovery systems. Policy-based replication supports a larger separation between the production and recovery system.
  • Determine the round-trip time (RTT) between the systems. The RTT can affect the throughput of replication since the storage arrays can transmit only a finite amount of data at a time. As the RTT increases, the maximum throughput decreases. For higher RTT, policy-based replication can achieve higher throughput than remote-copy function.
  • Determine the change rate of data for applications storing data on the system. When applications have high change rates with low bandwidth, the system can fail to achieve low RPOs.
  • Consider the cost of implementing policy-based replication and maintaining the recovery system and compare this with the cost of downtime during disaster recovery.
  • Policy-based replication can be configured between systems with lower-bandwidth links, but lower throughput results in a higher recovery point and may exceeds the set RPO value.
  • The use of policy-based replication protects data from a large-scale disaster to the site or to storage arrays. However, it does not provide complete protection against data corruption or cyberattack. In some cases, it can prevent data corruption if the corruption is not replicated to the recovery systems.
    Note: You can use FlashCopy and Safeguarded Copy function with policy-based replication to achieve protection against data corruption and cyberattack.
  • To use policy-based replication on the system requires a set amount of allocated I/O group memory. If you are currently using remote copy functions such as Global Mirror, Metro Mirror, or HyperSwap, you might need to reallocate I/O group memory to use policy-based replication. The system supports converting Global Mirror configuration to policy-based replication sites, which might also required adjustments to the current allocated memory for these features. For more information, see Bitmap space configuration.
  • IBM Storage Virtualize for Public Cloud on Microsoft Azure does not support policy-based replication on Azure Standard_D16s_v3 virtual machine type. Policy-based replication requires either a Standard_D32s_v3 or Standard D64_v3 virtual machine types. For more information, see Planning Azure virtual machines (VM).
  • IBM Storage Virtualize for Public Cloud on Amazon web services (AWS) supports policy-based replication only on c5.18x large platform (144GB memory). For more information about c5 EC2 instance models, see Planning EC2 instance type.
  • The following ports are required to be open on the system management IP address to allow the systems to replicate:
    • Fibre Channel and IP replication require port 7443 for REST API access for replication management.
    • Additionally, IP replication requires ports 3260 and 3265.
The following configuration limits and restrictions apply to policy-based replication:
  • The name of a volume group cannot be changed while a replication policy is assigned.
  • The name of a volume cannot be changed while the volume is in a volume group with a replication policy assigned.
  • Ownership groups are not supported with policy-based replication.
  • Policy-based replication is not supported on HyperSwap topology systems.
  • Policy-based replication cannot be used with volumes that are:
    • Image mode
    • HyperSwap
    • Configured to use Transparent Cloud Tiering (TCT)
    • VMware vSphere Virtual Volumes (vVols)
  • The following actions cannot be performed on a volume while the volume is in a volume group with a replication policy assigned:
    • Resize (expand or shrink)
    • Migrate to image mode, or add an image mode copy
    • Move to a different I/O group