I am trying to use NiFi's PutCassandraRecord to input data into a Cassandra cluster of 3 nodes(provided all 3 nodes address to PutCassandraRecord). The throughput I am getting isn't enough for my workload, so I tested using 3 different PutCassandraRecord processors and providing 1 node address to each processor. I was able to get greater throughput this route.
My question is could I split the amount of incoming flowfiles evenly to 3 or more different processors? More specifically, I am using ListenHTTP to listen for incoming data, and I want to split the incoming data evenly to 3 different PutCassandraRecord processors
I am running NiFi 1.9.2