Two Elasticsearch database servers use bidirectional asynchronous data replication Both servers accept writes from clients...

Two Elasticsearch database servers use bidirectional asynchronous data replication Both servers accept writes from clients The design must meet these requirements:

• The cluster must survive if a fault occurs that causes the network connection to go down between nodes

• The data must remain consistent if communication between nodes fails.

• The data must be spread evenly across all nodes in the cluster.

Which design approach must be used to meet the requirements?