mirror of
https://github.com/apache/impala.git
synced 2026-01-22 09:01:58 -05:00
04d027df13
It was disabled for performance reasons (IMPALA-1003) and this patch re-enables it since a lot of codegen improvements have happened since then. This patch switches the aggregation to use the CASE conditional instead of IF since the former has proper codegen support (IMPALA-7655). Tests: ===== - Updated the affected tests to include the null counts. - Added unit tests that verify IS [NOT] NULL predicates' cardinality estimation. Perf note: ========= I reran the compute stats child query with null counts included on the store_sales table from 1000 SF (1TB) tpcds dataset. The table had 22 non-partitioned columns (on which null counts were computed) and ~2.8B rows. This experiment showed around 7-8% perf drop compared to the same child query without null counts for these columns. Change-Id: Ic68f8b4c3756eb1980ce299a602a7d56db1e507a Reviewed-on: http://gerrit.cloudera.org:8080/11565 Reviewed-by: Impala Public Jenkins <impala-public-jenkins@cloudera.com> Tested-by: Impala Public Jenkins <impala-public-jenkins@cloudera.com>
65 lines
1.8 KiB
Plaintext
65 lines
1.8 KiB
Plaintext
====
|
|
---- QUERY
|
|
# test compute stats on a partitioned decimal text table
|
|
create table decimal_tbl like functional.decimal_tbl;
|
|
insert into decimal_tbl partition(d6)
|
|
select * from functional.decimal_tbl;
|
|
====
|
|
---- QUERY
|
|
compute stats decimal_tbl
|
|
---- RESULTS
|
|
'Updated 1 partition(s) and 5 column(s).'
|
|
====
|
|
---- QUERY
|
|
show table stats decimal_tbl
|
|
---- LABELS
|
|
d6, #Rows, #Files, Size, Bytes Cached, Cache Replication, Format, Incremental Stats, Location
|
|
---- RESULTS
|
|
'1',5,1,'375B','NOT CACHED','NOT CACHED','TEXT','false',regex:.*
|
|
'Total',5,1,'375B','0B','','','',''
|
|
---- TYPES
|
|
STRING, BIGINT, BIGINT, STRING, STRING, STRING, STRING, STRING, STRING
|
|
====
|
|
---- QUERY
|
|
show column stats decimal_tbl
|
|
---- LABELS
|
|
COLUMN, TYPE, #DISTINCT VALUES, #NULLS, MAX SIZE, AVG SIZE
|
|
---- RESULTS
|
|
'd1','DECIMAL(9,0)',4,0,4,4
|
|
'd2','DECIMAL(10,0)',3,0,8,8
|
|
'd3','DECIMAL(20,10)',5,0,16,16
|
|
'd4','DECIMAL(38,38)',1,0,16,16
|
|
'd5','DECIMAL(10,5)',5,0,8,8
|
|
'd6','DECIMAL(9,0)',1,0,4,4
|
|
---- TYPES
|
|
STRING, STRING, BIGINT, BIGINT, BIGINT, DOUBLE
|
|
====
|
|
---- QUERY
|
|
# test compute stats on a mixed-type parquet table
|
|
create table mixed_types(a int, b decimal(10,0)) stored as parquet;
|
|
insert into mixed_types values (1, 2), (3, 4);
|
|
====
|
|
---- QUERY
|
|
compute stats mixed_types
|
|
---- RESULTS
|
|
'Updated 1 partition(s) and 2 column(s).'
|
|
====
|
|
---- QUERY
|
|
show table stats mixed_types
|
|
---- LABELS
|
|
#Rows, #Files, Size, Bytes Cached, Cache Replication, Format, Incremental Stats, Location
|
|
---- RESULTS
|
|
2,1,regex:.+B,'NOT CACHED','NOT CACHED','PARQUET','false',regex:.*
|
|
---- TYPES
|
|
BIGINT, BIGINT, STRING, STRING, STRING, STRING, STRING, STRING
|
|
====
|
|
---- QUERY
|
|
show column stats mixed_types
|
|
---- LABELS
|
|
COLUMN, TYPE, #DISTINCT VALUES, #NULLS, MAX SIZE, AVG SIZE
|
|
---- RESULTS
|
|
'a','INT',2,0,4,4
|
|
'b','DECIMAL(10,0)',2,0,8,8
|
|
---- TYPES
|
|
STRING, STRING, BIGINT, BIGINT, BIGINT, DOUBLE
|
|
==== |