Skip to content

Latest commit

 

History

History
24 lines (19 loc) · 772 Bytes

README.md

File metadata and controls

24 lines (19 loc) · 772 Bytes

hadoop-fileformat-benchmark-kit


Aims to be a set of utilities to assist benchmarking performance for different fileformats for a given workload (Hive/Impala). Attributes it cares about -

  1. Size of blocks file
  2. Compression Ratio
  3. Query Performance - pending item

Warning this is a work in progress. At the moment, it does conversions for single tables using scripts

Usage


$ ./generate-conversion-hql.sh <input-db>.<input-table> <output-table-prefix> \
  > hive-bechmark.hql
$ hive -f hive-bechmark.hql

Known Issues


  • Avro conversion is not working at the moment

References