mirror of
https://github.com/apache/impala.git
synced 2026-02-02 06:00:36 -05:00
This script compares 2 runs of TPCDS by parsing their respective Impala plain text query profiles. It currently outputs the peak memory comparision of both runs where: 1. It compares average per-node peak memory and geo-mean per-node peak memory. 2. It compares max peak memory reduction among Hash operators. It can be extended to other comparisions in future. Example usage: tpcds_run_comparator.py <path to base run profile> <path to new run profile> [path to result csv file] Change-Id: Ib2e9ae1a2919156b0022072f47ff71d7775b20e6 Reviewed-on: http://gerrit.cloudera.org:8080/17855 Reviewed-by: Riza Suminto <riza.suminto@cloudera.com> Reviewed-by: Zoltan Borok-Nagy <boroknagyz@cloudera.com> Tested-by: Zoltan Borok-Nagy <boroknagyz@cloudera.com>