mirror of
https://github.com/apache/impala.git
synced 2026-01-27 06:10:53 -05:00
de7f09d726
This commit adds the command line flag enable_parquet_page_index_writing to the Impala daemon that switches Impala's ability of writing the Parquet page index. By default the flag is false, i.e. Impala doesn't write the page index. This flag is only temporary, we plan to remove it once Impala is able to read the page index and has better testing around it. Because of this change I had to move test_parquet_page_index.py to the custom_cluster test suite since I need to set this command line flag in order to test the functionality. I also merged most of the test cases because we don't want to restart the cluster too many times. I removed 'num_data_pages_' from BaseColumnWriter since it was rather confusing and didn't provide any measurable performance improvement. This commit fixes the ASAN error produced by the first IMPALA-7644 commit which was reverted later. Change-Id: Ib4a9098a2085a385351477c715ae245d83bf1c72 Reviewed-on: http://gerrit.cloudera.org:8080/11694 Reviewed-by: Impala Public Jenkins <impala-public-jenkins@cloudera.com> Tested-by: Impala Public Jenkins <impala-public-jenkins@cloudera.com>
200 lines
7.6 KiB
Plaintext
200 lines
7.6 KiB
Plaintext
====
|
|
---- QUERY
|
|
# This test relies on a deterministic row order so we use "sort by (id)".
|
|
create table alltypes sort by (id) like functional_parquet.alltypes;
|
|
alter table alltypes set tblproperties("impala.enable.stats.extrapolation"="true");
|
|
insert into alltypes partition(year, month)
|
|
select * from functional_parquet.alltypes where year = 2009;
|
|
====
|
|
---- QUERY
|
|
# No stats are available.
|
|
explain select id from alltypes;
|
|
---- RESULTS: VERIFY_IS_SUBSET
|
|
' stored statistics:'
|
|
' table: rows=unavailable size=unavailable'
|
|
' partitions: 0/12 rows=unavailable'
|
|
' columns: unavailable'
|
|
row_regex:.* extrapolated-rows=unavailable.*
|
|
' tuple-ids=0 row-size=4B cardinality=unavailable'
|
|
---- TYPES
|
|
STRING
|
|
====
|
|
---- QUERY
|
|
compute stats alltypes
|
|
---- RESULTS
|
|
'Updated 1 partition(s) and 11 column(s).'
|
|
---- TYPES
|
|
STRING
|
|
====
|
|
---- QUERY
|
|
# Only the table-level row count is stored. The partition row counts are extrapolated.
|
|
show table stats alltypes
|
|
---- LABELS
|
|
YEAR, MONTH, #ROWS, EXTRAP #ROWS, #FILES, SIZE, BYTES CACHED, CACHE REPLICATION, FORMAT, INCREMENTAL STATS, LOCATION
|
|
---- RESULTS
|
|
'2009','1',-1,308,1,regex:.*B,'NOT CACHED','NOT CACHED','PARQUET','false','$NAMENODE/test-warehouse/$DATABASE.db/alltypes/year=2009/month=1'
|
|
'2009','2',-1,288,1,regex:.*B,'NOT CACHED','NOT CACHED','PARQUET','false','$NAMENODE/test-warehouse/$DATABASE.db/alltypes/year=2009/month=2'
|
|
'2009','3',-1,308,1,regex:.*B,'NOT CACHED','NOT CACHED','PARQUET','false','$NAMENODE/test-warehouse/$DATABASE.db/alltypes/year=2009/month=3'
|
|
'2009','4',-1,302,1,regex:.*B,'NOT CACHED','NOT CACHED','PARQUET','false','$NAMENODE/test-warehouse/$DATABASE.db/alltypes/year=2009/month=4'
|
|
'2009','5',-1,308,1,regex:.*B,'NOT CACHED','NOT CACHED','PARQUET','false','$NAMENODE/test-warehouse/$DATABASE.db/alltypes/year=2009/month=5'
|
|
'2009','6',-1,302,1,regex:.*B,'NOT CACHED','NOT CACHED','PARQUET','false','$NAMENODE/test-warehouse/$DATABASE.db/alltypes/year=2009/month=6'
|
|
'2009','7',-1,308,1,regex:.*B,'NOT CACHED','NOT CACHED','PARQUET','false','$NAMENODE/test-warehouse/$DATABASE.db/alltypes/year=2009/month=7'
|
|
'2009','8',-1,308,1,regex:.*B,'NOT CACHED','NOT CACHED','PARQUET','false','$NAMENODE/test-warehouse/$DATABASE.db/alltypes/year=2009/month=8'
|
|
'2009','9',-1,302,1,regex:.*B,'NOT CACHED','NOT CACHED','PARQUET','false','$NAMENODE/test-warehouse/$DATABASE.db/alltypes/year=2009/month=9'
|
|
'2009','10',-1,308,1,regex:.*B,'NOT CACHED','NOT CACHED','PARQUET','false','$NAMENODE/test-warehouse/$DATABASE.db/alltypes/year=2009/month=10'
|
|
'2009','11',-1,302,1,regex:.*B,'NOT CACHED','NOT CACHED','PARQUET','false','$NAMENODE/test-warehouse/$DATABASE.db/alltypes/year=2009/month=11'
|
|
'2009','12',-1,308,1,regex:.*B,'NOT CACHED','NOT CACHED','PARQUET','false','$NAMENODE/test-warehouse/$DATABASE.db/alltypes/year=2009/month=12'
|
|
'Total','',3650,3650,12,regex:.*B,'0B','','','',''
|
|
---- TYPES
|
|
STRING,STRING,BIGINT,BIGINT,BIGINT,STRING,STRING,STRING,STRING,STRING,STRING
|
|
====
|
|
---- QUERY
|
|
# Stats are available now.
|
|
explain select id from alltypes;
|
|
---- RESULTS: VERIFY_IS_EQUAL
|
|
row_regex:.*Max Per-Host Resource Reservation: Memory=.*
|
|
row_regex:.*Per-Host Resource Estimates: Memory=.*
|
|
'Codegen disabled by planner'
|
|
''
|
|
'F00:PLAN FRAGMENT [UNPARTITIONED] hosts=1 instances=1'
|
|
row_regex:.*Per-Host Resources: mem-estimate=.* mem-reservation=.*
|
|
'PLAN-ROOT SINK'
|
|
row_regex:.*mem-estimate=.* mem-reservation=.*
|
|
'|'
|
|
'00:SCAN HDFS [$DATABASE.alltypes]'
|
|
row_regex:.*partitions=12/12 files=12 size=.*
|
|
' stored statistics:'
|
|
row_regex:.*table: rows=3650 size=.*
|
|
' partitions: 0/12 rows=unavailable'
|
|
' columns: all'
|
|
row_regex:.* extrapolated-rows=3650.*
|
|
row_regex:.*mem-estimate=.* mem-reservation=.*
|
|
' tuple-ids=0 row-size=4B cardinality=3650'
|
|
' in pipelines: 00(GETNEXT)'
|
|
---- TYPES
|
|
STRING
|
|
====
|
|
---- QUERY
|
|
# Select a subset of partitions.
|
|
explain select id from alltypes where month in (1, 2, 3);
|
|
---- RESULTS: VERIFY_IS_EQUAL
|
|
row_regex:.*Max Per-Host Resource Reservation: Memory=.*
|
|
row_regex:.*Per-Host Resource Estimates: Memory=.*
|
|
'Codegen disabled by planner'
|
|
''
|
|
'F00:PLAN FRAGMENT [UNPARTITIONED] hosts=1 instances=1'
|
|
row_regex:.*mem-estimate=.* mem-reservation=.*
|
|
'PLAN-ROOT SINK'
|
|
row_regex:.*mem-estimate=.* mem-reservation=.*
|
|
'|'
|
|
'00:SCAN HDFS [$DATABASE.alltypes]'
|
|
row_regex:.*partitions=3/12 files=3 size=.*
|
|
' stored statistics:'
|
|
row_regex:.*table: rows=3650 size=.*
|
|
' partitions: 0/3 rows=unavailable'
|
|
' columns: all'
|
|
row_regex:.* extrapolated-rows=904.*
|
|
row_regex:.*mem-estimate=.* mem-reservation=.*
|
|
' tuple-ids=0 row-size=4B cardinality=904'
|
|
' in pipelines: 00(GETNEXT)'
|
|
---- TYPES
|
|
STRING
|
|
====
|
|
---- QUERY
|
|
# Double the data in existing partitions.
|
|
insert into alltypes partition(year, month)
|
|
select * from functional_parquet.alltypes where year = 2009;
|
|
explain select id from alltypes;
|
|
---- RESULTS: VERIFY_IS_EQUAL
|
|
row_regex:.*Max Per-Host Resource Reservation: Memory=.*
|
|
row_regex:.*Per-Host Resource Estimates: Memory=.*
|
|
''
|
|
'F00:PLAN FRAGMENT [UNPARTITIONED] hosts=1 instances=1'
|
|
row_regex:.*Per-Host Resources: mem-estimate=.* mem-reservation=.*
|
|
'PLAN-ROOT SINK'
|
|
row_regex:.*mem-estimate=.* mem-reservation=.*
|
|
'|'
|
|
'00:SCAN HDFS [$DATABASE.alltypes]'
|
|
row_regex:.*partitions=12/12 files=24 size=.*
|
|
' stored statistics:'
|
|
row_regex:.*table: rows=3650 size=.*
|
|
' partitions: 0/12 rows=unavailable'
|
|
' columns: all'
|
|
row_regex:.* extrapolated-rows=7300.*
|
|
row_regex:.*mem-estimate=.* mem-reservation=.*
|
|
' tuple-ids=0 row-size=4B cardinality=7300'
|
|
' in pipelines: 00(GETNEXT)'
|
|
---- TYPES
|
|
STRING
|
|
====
|
|
---- QUERY
|
|
# Create new partitions and extrapolate their row count.
|
|
insert into alltypes partition(year, month)
|
|
select * from functional_parquet.alltypes where year = 2010;
|
|
explain select id from alltypes where year = 2010;
|
|
---- RESULTS: VERIFY_IS_EQUAL
|
|
row_regex:.*Max Per-Host Resource Reservation: Memory=.*
|
|
row_regex:.*Per-Host Resource Estimates: Memory=16MB'
|
|
'Codegen disabled by planner'
|
|
''
|
|
'F00:PLAN FRAGMENT [UNPARTITIONED] hosts=1 instances=1'
|
|
row_regex:.*Per-Host Resources: mem-estimate=.* mem-reservation=.*
|
|
'PLAN-ROOT SINK'
|
|
row_regex:.*mem-estimate=.* mem-reservation=.*
|
|
'|'
|
|
'00:SCAN HDFS [$DATABASE.alltypes]'
|
|
row_regex:.*partitions=12/24 files=12 size=.*
|
|
' stored statistics:'
|
|
row_regex:.*table: rows=3650 size=.*
|
|
' partitions: 0/12 rows=unavailable'
|
|
' columns: all'
|
|
row_regex:.* extrapolated-rows=3651.*
|
|
row_regex:.*mem-estimate=.* mem-reservation=.*
|
|
' tuple-ids=0 row-size=4B cardinality=3651'
|
|
' in pipelines: 00(GETNEXT)'
|
|
---- TYPES
|
|
STRING
|
|
====
|
|
---- QUERY
|
|
# Compute stats and run the same query again.
|
|
compute stats alltypes;
|
|
explain select id from alltypes where year = 2010;
|
|
---- RESULTS: VERIFY_IS_EQUAL
|
|
row_regex:.*Max Per-Host Resource Reservation: Memory=.*
|
|
row_regex:.*Per-Host Resource Estimates: Memory=16MB'
|
|
'Codegen disabled by planner'
|
|
''
|
|
'F00:PLAN FRAGMENT [UNPARTITIONED] hosts=1 instances=1'
|
|
row_regex:.*Per-Host Resources: mem-estimate=.* mem-reservation=.*
|
|
'PLAN-ROOT SINK'
|
|
row_regex:.*mem-estimate=.* mem-reservation=.*
|
|
'|'
|
|
'00:SCAN HDFS [$DATABASE.alltypes]'
|
|
row_regex:.*partitions=12/24 files=12 size=.*
|
|
' stored statistics:'
|
|
row_regex:.*table: rows=10950 size=.*
|
|
' partitions: 0/12 rows=unavailable'
|
|
' columns: all'
|
|
row_regex:.* extrapolated-rows=3651
|
|
row_regex:.*mem-estimate=.* mem-reservation=.*
|
|
' tuple-ids=0 row-size=4B cardinality=3651'
|
|
' in pipelines: 00(GETNEXT)'
|
|
---- TYPES
|
|
STRING
|
|
====
|
|
---- QUERY
|
|
# Test that dropping stats resets everything.
|
|
drop stats alltypes;
|
|
explain select id from alltypes;
|
|
---- RESULTS: VERIFY_IS_SUBSET
|
|
' stored statistics:'
|
|
' table: rows=unavailable size=unavailable'
|
|
' partitions: 0/24 rows=unavailable'
|
|
' columns: unavailable'
|
|
row_regex:.* extrapolated-rows=unavailable.*
|
|
' tuple-ids=0 row-size=4B cardinality=unavailable'
|
|
' in pipelines: 00(GETNEXT)'
|
|
---- TYPES
|
|
STRING
|
|
====
|