In a deployment in which each target system receives data from a standalone instance of a target service, load balancing is not a requirement.
EDS duplicates the data across the data targets.
EDS delivers all the messages published by the source service to all the data target services so that they have the complete data set for analysis.
For example, you create separate target services to send data to an HDFS cluster, a Cassandra database, and a RulePoint instance.
The following image shows
EDS data duplication across the data targets:
EDS duplicates data as follows:
A source service reads data from the data source and publishes it as two messages over a topic called logs.
EDS delivers both the messages to all the target services.
The target services receive both messages and send them to the targets.