mirror of
https://github.com/apache/impala.git
synced 2026-02-01 21:00:29 -05:00
f932d78ad0
This patch provides count(star) optimization for ORC scans, similar to the work done in IMPALA-5036 for Parquet scans. We use the stripes num rows statistics when computing the count star instead of materializing empty rows. The aggregate function changed from a count to a special sum function initialized to 0. This count(star) optimization is disabled for the full ACID table because the scanner might need to read and validate the 'currentTransaction' column in table's special schema. This patch drops 'parquet' from names related to the count star optimization. It also improves the count(star) operation in general by serving the result just from the file's footer stats for both Parquet and ORC. We unify the optimized count star and zero slot scan functions into HdfsColumnarScanner. The following table shows a performance comparison before and after the patch. primitive_count_star query target tpch10_parquet.lineitem table (10GB scale TPC-H). Meanwhile, count_star_parq and count_star_orc query is a modified primitive_count_star query that targets tpch_parquet.lineitem and tpch_orc_def.lineitem table accordingly. +-------------------+----------------------+-----------------------+--------+-------------+------------+------------+----------------+-------+----------------+---------+-------+ | Workload | Query | File Format | Avg(s) | Base Avg(s) | Delta(Avg) | StdDev(%) | Base StdDev(%) | Iters | Median Diff(%) | MW Zval | Tval | +-------------------+----------------------+-----------------------+--------+-------------+------------+------------+----------------+-------+----------------+---------+-------+ | tpch_parquet | count_star_parq | parquet / none / none | 0.06 | 0.07 | -10.45% | 2.87% | * 25.51% * | 9 | -1.47% | -1.26 | -1.22 | | tpch_orc_def | count_star_orc | orc / def / none | 0.06 | 0.08 | -22.37% | 6.22% | * 30.95% * | 9 | -1.85% | -1.16 | -2.14 | | TARGETED-PERF(10) | primitive_count_star | parquet / none / none | 0.06 | 0.08 | I -30.40% | 2.68% | * 29.63% * | 9 | I -7.20% | -2.42 | -3.07 | +-------------------+----------------------+-----------------------+--------+-------------+------------+------------+----------------+-------+----------------+---------+-------+ Testing: - Add PlannerTest.testOrcStatsAgg - Add TestAggregationQueries::test_orc_count_star_optimization - Exercise count(star) in TestOrc::test_misaligned_orc_stripes - Pass core tests Change-Id: I0fafa1182f97323aeb9ee39dd4e8ecd418fa6091 Reviewed-on: http://gerrit.cloudera.org:8080/18327 Reviewed-by: Impala Public Jenkins <impala-public-jenkins@cloudera.com> Tested-by: Impala Public Jenkins <impala-public-jenkins@cloudera.com>
176 lines
3.4 KiB
Plaintext
176 lines
3.4 KiB
Plaintext
====
|
|
---- QUERY
|
|
# Basic partition key scan.
|
|
select distinct year
|
|
from alltypes
|
|
---- RESULTS
|
|
2009
|
|
2010
|
|
---- TYPES
|
|
INT
|
|
---- RUNTIME_PROFILE
|
|
# Confirm that only one row per file is read.
|
|
aggregation(SUM, RowsRead): 24
|
|
---- RUNTIME_PROFILE: table_format=parquet,orc
|
|
# Confirm that only one metadata per file is read.
|
|
aggregation(SUM, RowsRead): 0
|
|
aggregation(SUM, NumFileMetadataRead): 24
|
|
====
|
|
---- QUERY
|
|
# Test with more complex multiple distinct aggregation.
|
|
select count(distinct year), count(distinct month)
|
|
from alltypes
|
|
---- RESULTS
|
|
2,12
|
|
---- TYPES
|
|
BIGINT,BIGINT
|
|
---- RUNTIME_PROFILE
|
|
# Confirm that only one row per file is read.
|
|
aggregation(SUM, RowsRead): 24
|
|
---- RUNTIME_PROFILE: table_format=parquet,orc
|
|
# Confirm that only one metadata per file is read.
|
|
aggregation(SUM, RowsRead): 0
|
|
aggregation(SUM, NumFileMetadataRead): 24
|
|
====
|
|
---- QUERY
|
|
# Distinct aggregation with multiple columns.
|
|
select distinct year, month
|
|
from alltypes
|
|
---- RESULTS
|
|
2009,1
|
|
2009,2
|
|
2009,3
|
|
2009,4
|
|
2009,5
|
|
2009,6
|
|
2009,7
|
|
2009,8
|
|
2009,9
|
|
2009,10
|
|
2009,11
|
|
2009,12
|
|
2010,1
|
|
2010,2
|
|
2010,3
|
|
2010,4
|
|
2010,5
|
|
2010,6
|
|
2010,7
|
|
2010,8
|
|
2010,9
|
|
2010,10
|
|
2010,11
|
|
2010,12
|
|
---- TYPES
|
|
INT,INT
|
|
---- RUNTIME_PROFILE
|
|
# Confirm that only one row per file is read.
|
|
aggregation(SUM, RowsRead): 24
|
|
---- RUNTIME_PROFILE: table_format=parquet,orc
|
|
# Confirm that only one metadata per file is read.
|
|
aggregation(SUM, RowsRead): 0
|
|
aggregation(SUM, NumFileMetadataRead): 24
|
|
====
|
|
---- QUERY
|
|
# Partition key scan combined with analytic function.
|
|
select year, row_number() over (order by year)
|
|
from alltypes group by year;
|
|
---- RESULTS
|
|
2009,1
|
|
2010,2
|
|
---- TYPES
|
|
INT,BIGINT
|
|
---- RUNTIME_PROFILE
|
|
# Confirm that only one row per file is read.
|
|
aggregation(SUM, RowsRead): 24
|
|
---- RUNTIME_PROFILE: table_format=parquet,orc
|
|
# Confirm that only one metadata per file is read.
|
|
aggregation(SUM, RowsRead): 0
|
|
aggregation(SUM, NumFileMetadataRead): 24
|
|
====
|
|
---- QUERY
|
|
# Partition scan combined with sort.
|
|
select distinct year, month
|
|
from alltypes
|
|
order by year, month
|
|
---- RESULTS
|
|
2009,1
|
|
2009,2
|
|
2009,3
|
|
2009,4
|
|
2009,5
|
|
2009,6
|
|
2009,7
|
|
2009,8
|
|
2009,9
|
|
2009,10
|
|
2009,11
|
|
2009,12
|
|
2010,1
|
|
2010,2
|
|
2010,3
|
|
2010,4
|
|
2010,5
|
|
2010,6
|
|
2010,7
|
|
2010,8
|
|
2010,9
|
|
2010,10
|
|
2010,11
|
|
2010,12
|
|
---- TYPES
|
|
INT,INT
|
|
---- RUNTIME_PROFILE
|
|
# Confirm that only one row per file is read.
|
|
aggregation(SUM, RowsRead): 24
|
|
---- RUNTIME_PROFILE: table_format=parquet,orc
|
|
# Confirm that only one metadata per file is read.
|
|
aggregation(SUM, RowsRead): 0
|
|
aggregation(SUM, NumFileMetadataRead): 24
|
|
====
|
|
---- QUERY
|
|
# Partition key scan combined with predicate on partition columns
|
|
select distinct year, month
|
|
from alltypes
|
|
where year - 2000 = month;
|
|
---- RESULTS
|
|
2009,9
|
|
2010,10
|
|
---- TYPES
|
|
INT,INT
|
|
---- RUNTIME_PROFILE
|
|
# Confirm that only one row per file is read.
|
|
aggregation(SUM, RowsRead): 2
|
|
---- RUNTIME_PROFILE: table_format=parquet,orc
|
|
# Confirm that only one metadata per file is read.
|
|
aggregation(SUM, RowsRead): 0
|
|
aggregation(SUM, NumFileMetadataRead): 2
|
|
====
|
|
---- QUERY
|
|
# Partition key scan combined with having predicate.
|
|
select year, min(month)
|
|
from alltypes
|
|
group by year
|
|
having min(month) = 1
|
|
---- RESULTS
|
|
2009,1
|
|
2010,1
|
|
---- TYPES
|
|
INT,INT
|
|
---- RUNTIME_PROFILE
|
|
# Confirm that only one row per file is read.
|
|
aggregation(SUM, RowsRead): 24
|
|
---- RUNTIME_PROFILE: table_format=parquet,orc
|
|
# Confirm that only one metadata per file is read.
|
|
aggregation(SUM, RowsRead): 0
|
|
aggregation(SUM, NumFileMetadataRead): 24
|
|
====
|
|
---- QUERY
|
|
# Empty table should not return any rows
|
|
select distinct 'test'
|
|
from emptytable
|
|
---- RESULTS
|
|
---- TYPES
|
|
STRING
|
|
====
|