I have noticed a very low throughput in our pipeline when inserting data from spark to cassandra (less than 1 MB/s per core). When trying to tune write conf (spark.cas
spark.cas