Files
impala/testdata/workloads/functional-query/queries/QueryTest/mixed-format.test
Riza Suminto f932d78ad0 IMPALA-11123: Optimize count(star) for ORC scans
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>
2022-04-05 13:27:10 +00:00

42 lines
1.2 KiB
Plaintext

====
---- QUERY
# Check that data from mixed format partitions can be read
# transparently. Compute sum in order to force deserialisation
select count(*), sum(int_col) from functional.alltypesmixedformat
---- TYPES
bigint, bigint
---- RESULTS
1200,5400
====
---- QUERY
# Restrict set of partitions (still multi-format)
select count(*), sum(int_col) from functional.alltypesmixedformat where month = 1 or month = 3
---- TYPES
bigint, bigint
---- RESULTS
620,2790
====
---- QUERY
# Read single partition alone
select count(*), sum(int_col) from functional.alltypesmixedformat where month = 2
---- TYPES
bigint, bigint
---- RESULTS
280,1260
====
---- QUERY
# IMPALA-11123: IMPALA-5861 add this test to verify that 'RowRead' counter is not double
# counted for zero slot scan. IMPALA-11123 remove incerement of 'RowRead' counter
# in case of optimized count(star) and zero slot scan query. This cause reduction of
# 'RowsRead' value from 1200 to 900 since the other 300 are served through
# zero slot scan. We do not verify 'NumFileMetadataRead' since it does not stay the same
# over different test vector permutation.
select count(*) from functional.alltypesmixedformat
---- TYPES
bigint
---- RESULTS
1200
---- RUNTIME_PROFILE
aggregation(SUM, RowsRead): 900
====