Performance Benchmark: Business Intelligence on Big Data Q4 '16

As more enterprises run BI on Hadoop, a critical question arises: "How 'fit' are SQL-on-Hadoop engines to satisfy BI workloads?"

In the first study of its kind, we answer that question --- with performance measures derived from real-world testing on actual workloads --- across the latest engines: Impala, Spark, Presto, and Hive.

Our evaluation looks at how well these engines satisfy three key criteria:

  • Performs on Big Data
  • Fast on Small Data
  • Stable for Many Users

...with the latest data updated for Fall 2016.