The TPC Benchmark* Express-BigBench (TPCx-BB) measures the performance of end-to-end big data analytics by implementing 30 use cases that simulate big data processing, big data analytics, and reporting. The data represented are either structured, semi-structured, or un-structured data types. These use cases are frequently performed by big data operations at retailers with both physical and online store presence.
The TPCx-BB performance metric is called the Big Bench Query-per-minute (BBQpm@Size), where size is the scale factor of the data. The metric reflects three test phases: a load test that aggregates data from various sources and formats; a power test that runs each use case once to identify optimization areas and utilization patterns; and a throughput test that runs multiple jobs in parallel to test the efficiency of the cluster. TPCx-BB is implemented to work with modern big data analytics frameworks residing in the Hadoop* ecosystem such as Map Reduce*, Hive*, Spark*, Tez*, and MLLIB*.