About replication
This section describes how to configure replication on the Core. You can configure Cores for seamless failover and recovery between data centers without any data loss. In this environment, your primary and secondary data centers are always synchronized in case of large scale failures such as power loss, natural disasters, or hardware failure. You connect two separate Cores that are each connected to their own storage array in two separate data centers.
The primary data center receives all the reads and writes from the Edge, and is connected to the secondary data center. The secondary data center hosts the replica LUNs, which are copies of the primary LUNs located at the primary data center.
Replication can interoperate with all existing Cores and Edges as long as the software version is compatible).
You must meet these requirements to set up replication:
• The backend storage array must be configured for each Core that will be included in the replication configuration.
• The primary data center should be able to reach the secondary data center through the chosen interfaces.
• The secondary Core can’t have any Edges or LUNs.
• Each Core must have the same configuration. For example, if high availability is configured on the primary data center’s Cores, it must also be configured on the secondary Cores.
• The Edges should be able to reach the secondary data center.
Before you begin
1. Create discoverable LUNs on storage arrays in both data centers. The LUNs you create on the primary data center will serve the Edges during normal operation. The LUNs you create on the secondary data center are replicas of the primary data center LUNs and should not be added to the Core. The secondary data center LUNs should be exactly the same size as those in the primary data center so they can serve the Edges in the event of a primary data center failure.
Storage array vendors and models can vary between data centers, as well as LUN size. For this reason, we permit a size leeway of 1 GiB to allow the LUN created on the secondary data center storage array to be at most 1 GiB larger (but not smaller) than the primary data center LUN. For example, a 4 GiB LUN on the primary data center may have a 5 GiB replica LUN on the secondary data center. The larger part of the LUN in the secondary data center isn’t used.
2. Create the Journal LUN you will be using on the backend. The Journal LUN is a dedicated LUN on the backend storage that is used to temporarily journal writes for replicated LUNs when replication is suspended. If you have Cores set up for high availability, share the same Journal LUN between the Cores in case one of the Cores fails.
The sizing of the Journal LUN is dependent on the number of LUNs you want to replicate. We recommend a thinly provisioned LUN of 500 GiB or more. If the Journal LUN isn’t large enough, an alarm will be triggered with a reminder of the minimum required size (calculated based on the currently configured replica LUNs). There is no maximum size for the Journal LUN.
3. Ensure that the Journal LUN is accessible by the Core, and that the Journal LUN isn’t added to the configuration.
4. If you plan to configure Cores for high availability, ensure that they are set up before you start configuring replication.