Robin Moffatt • 12/20/2016

ETL Offload with Spark and Amazon EMR - Part 4 - Analysing the Data

This article, part 4 of a series, details the analysis phase of a Spark/EMR ETL project. It evaluates 'SQL-on-Hadoop' engines (e.g., Apache Drill, Hive, Presto) for querying data stored in open formats like Parquet on S3/HDFS. The analysis compares performance, ANSI SQL support, and operational complexity against traditional RDBMS, highlighting the flexibility of decoupled storage and compute.

0 comments

#data analysis #Spark #Etl