I have developed my application using Spark + Hbase and Phoenix. I want to optimize the operations cost. Therefore I was looking at enhanced reliability for Apache HBase applications using MapR M7 (https://aws.amazon.com/emr/mapr/
). Still not convinced why I need MapR. I have deployed number of projects using Cloudera, but since we are using S3 it make sense to use EMR. Using Hbase with Cloudera we did not encounter many issues. Therefore trying to figure out why we need MapR.