3 documented AWS EMR cost inefficiencies, each with how to detect it and how to fix it.
AWS EMRCER-0057
Amazon EMR clusters often run on large, multi-node EC2 fleets, making them costly to leave running unnecessarily. If a cluster becomes idle-no longer processing jobs - but is not terminated, it continues accruing EC2 and EMR service...
AWS EMRCER-0395
Amazon EMR task nodes run processing work but do not store persistent data in HDFS. If a task node running on a Spot Instance is interrupted, no data is lost and the effect on the cluster is minimal, which is why the EMR guidelines list...
AWS EMRCER-0396
Long-running Amazon EMR clusters are often provisioned with a fixed number of core and task nodes sized for the heaviest job or the busiest time of day. Between peaks, all of those nodes keep running and billing even when YARN has little...
