When configuring Shareplex instance with named queues, it may be worth knowing if any limitation exists as to the # of named queues that can be created in a Shareplex instance, the reason why the # may reach high, and the workaround, if any, in case a limit exists. This solution delves on these.
Shareplex only allows a maximum of 25 queues per Shareplex instance (sp_cop) which may include named queues.
The following are the reasons for running into a very large # of queues in a Shareplex instance:
The same Shareplex instance acts as a source as well as target. This could be due to a high availability or a peer-to-peer replication environment. Consequently you may see more number of queues than normal.
Tables are segregated (grouped) queue wise for better control and avoidance of certain errors. If the # of tables in replication is very large and/or a very large number of groups are created, this can result in reaching this limit of 25 named queues.
If the # of targets are many, then this combined with a high # of tables can also result in very large # of named queues. This will be contingent upon named Export queues created target wise. If only one queue is configured per target, then this limitation will not arise.
The only way to get around this limitation of 25 queues per Shareplex instance is to use multiple instances of Shareplex to replicate between same set of source and target, with each instance replicating a set of tables. However, this may result in more resource usage and more work to manage replication.
Check Shareplex Administrator Guide Chapter titled "Preparing for Shareplex Replication" section titled "Configuration for isolating data using named export queues" for further details on the limitation only.