Designed and developed data flow with Cardinal Health Incubator Group (Fuse) to deliver realtime data from Healthcare EDI Records parsed and reported through Apache Hive into QlikView. Mirth listening endpoints stored files into Amazon S3 buckets for metadata partitioning. My contribution was to develop the flow in bringing in data required for daily reporting from pharmaceutical manufacturers. The data was partitioned by date, allowing the resource consumption for maintining the billions of records in each fact table to be relatively small, e.g. Hive was (8gb *4core) x 1 name + 2 data node cluster on Amazon.