In the intro to chapter 14, it mentions replication first and then shared store.

~~~
There are two main High Availability (HA) policies available, each deploying a different strategy to enable HA. The first, replication, synchronizes the data between master and slave brokers over the network. The second HA policy, shared store, involves the master and slave brokers sharing a location for their messaging data.
~~~

From discussions with engineering, it seems "shared store" is the preferred approach to high availability.

Can this preferred approach stated somewhere in the documentation so customer can prioritise their approach? Right now, I think most people will go for the replication approach, simply as it appears in the doc first and there is no "preferred approach" stated.