Alex Behm has uploaded a new patch set (#2). Change subject: IMPALA-4055: Speed up to_date() with custom implementation. ......................................................................
IMPALA-4055: Speed up to_date() with custom implementation. Simple implementation of to_date() that avoids calling into boost for a speedup of 10x. Perf: I generated a synthetic Parquet table with 26437248 rows and a single timestamp column. I tested the response time of the following query before and after this change. set mt_dop=1; select count(*) from to_date_test where to_date(ts) = '2017-10-23'; Before: 38.1s After: 3.3s Testing: I locally ran expr-test.cc and expr_test.py. Change-Id: I5713b3e0c27b739aae597a6911cf3b2ddd01f822 --- M be/src/exprs/timestamp-functions-ir.cc M be/src/exprs/timestamp-functions.cc M be/src/exprs/timestamp-functions.h 3 files changed, 29 insertions(+), 12 deletions(-) git pull ssh://gerrit.cloudera.org:29418/Impala-ASF refs/changes/91/5791/2 -- To view, visit http://gerrit.cloudera.org:8080/5791 To unsubscribe, visit http://gerrit.cloudera.org:8080/settings Gerrit-MessageType: newpatchset Gerrit-Change-Id: I5713b3e0c27b739aae597a6911cf3b2ddd01f822 Gerrit-PatchSet: 2 Gerrit-Project: Impala-ASF Gerrit-Branch: master Gerrit-Owner: Alex Behm <alex.b...@cloudera.com>
