-
Bug
-
Resolution: Done
-
Major
-
6.3, 6.2.1
-
None
-
Sprint 2017-10, Sprint 2017-11, Sprint 2017-12, Sprint 2017-13
-
1
-
"Load-Balancing Cluster Toplogy is ideal for load balancing non-persistent messages across brokers and for providing high-availability." This is mentioned at 7.3.1. Load-Balancing Cluster, please refer doc[1].
We are not able to understand why it is just mentioned for non-persistent messages. We believe that we should also append more information so that it is more useful for all of us.
More information should be:
This topology is most useful if one have a scenario where producer is generating a heavy load but do not really care if all messages make it to the consumer. This tends to the scenario where non persistent messages are used. As this topology doesn't have shared storage so If one is looking for better guarantees regarding delivery, probably best to combine networks and master slave as it is defined in subsequent sections.
