mirror of
https://github.com/apache/impala.git
synced 2026-01-22 18:02:34 -05:00
8cbec20ea4
Added rough estimates for exchange node and a justification of the method in the in-line comments. Testing: Updated Planner tests. Change-Id: I5b577f9511abc48b992e814d50bba4959f23f7fd Reviewed-on: http://gerrit.cloudera.org:8080/11692 Reviewed-by: Impala Public Jenkins <impala-public-jenkins@cloudera.com> Tested-by: Impala Public Jenkins <impala-public-jenkins@cloudera.com>
71 lines
3.2 KiB
Plaintext
71 lines
3.2 KiB
Plaintext
====
|
|
---- QUERY
|
|
# Explain a simple hash join query.
|
|
explain
|
|
select *
|
|
from tpch.lineitem join tpch.orders on l_orderkey = o_orderkey;
|
|
---- RESULTS: VERIFY_IS_EQUAL
|
|
row_regex:.*Max Per-Host Resource Reservation: Memory=[0-9.]*MB Threads=[0-9]*.*
|
|
row_regex:.*Per-Host Resource Estimates: Memory=[0-9.]*MB.*
|
|
''
|
|
'F02:PLAN FRAGMENT [UNPARTITIONED] hosts=1 instances=1'
|
|
row_regex:.*| Per-Host Resources: mem-estimate=[0-9.]*[A-Z]*B mem-reservation=[0-9.]*[A-Z]*B thread-reservation=1
|
|
'PLAN-ROOT SINK'
|
|
row_regex:.*| mem-estimate=[0-9.]*[A-Z]*B mem-reservation=[0-9.]*[A-Z]*B thread-reservation=0
|
|
'|'
|
|
'04:EXCHANGE [UNPARTITIONED]'
|
|
row_regex:.*| mem-estimate=[0-9.]*[A-Z]*B mem-reservation=[0-9.]*[A-Z]*B thread-reservation=0
|
|
'| tuple-ids=0,1 row-size=454B cardinality=5757710'
|
|
'| in pipelines: 00(GETNEXT)'
|
|
'|'
|
|
'F00:PLAN FRAGMENT [RANDOM] hosts=3 instances=3'
|
|
row_regex:.*Per-Host Resources: mem-estimate=[0-9.]*[A-Z]*B mem-reservation=[0-9.]*[A-Z]*B thread-reservation=.*
|
|
'02:HASH JOIN [INNER JOIN, BROADCAST]'
|
|
'| hash predicates: l_orderkey = o_orderkey'
|
|
'| fk/pk conjuncts: l_orderkey = o_orderkey'
|
|
'| runtime filters: RF000[bloom] <- o_orderkey'
|
|
row_regex:.*| mem-estimate=[0-9.]*[A-Z]*B mem-reservation=[0-9.]*[A-Z]*B spill-buffer=[0-9.]*MB thread-reservation=0.*
|
|
'| tuple-ids=0,1 row-size=454B cardinality=5757710'
|
|
'| in pipelines: 00(GETNEXT), 01(OPEN)'
|
|
'|'
|
|
'|--03:EXCHANGE [BROADCAST]'
|
|
row_regex:.*| | mem-estimate=[0-9.]*[A-Z]*B mem-reservation=[0-9.]*[A-Z]*B thread-reservation=0
|
|
'| | tuple-ids=1 row-size=191B cardinality=1500000'
|
|
'| | in pipelines: 01(GETNEXT)'
|
|
'| |'
|
|
'| F01:PLAN FRAGMENT [RANDOM] hosts=2 instances=2'
|
|
row_regex:.*| Per-Host Resources: mem-estimate=[0-9.]*[A-Z]*B mem-reservation=[0-9.]*[A-Z]*B thread-reservation=.*
|
|
'| 01:SCAN HDFS [tpch.orders, RANDOM]'
|
|
row_regex:.*partitions=1/1 files=1 size=.*
|
|
'| stored statistics:'
|
|
row_regex:.*table: rows=1500000 size=.*
|
|
'| columns: all'
|
|
row_regex:.*| extrapolated-rows=disabled max-scan-range-rows=[0-9]*.*
|
|
row_regex:.*| mem-estimate=[0-9.]*[A-Z]*B mem-reservation=[0-9.]*[A-Z]*B thread-reservation=1.*
|
|
'| tuple-ids=1 row-size=191B cardinality=1500000'
|
|
'| in pipelines: 01(GETNEXT)'
|
|
'|'
|
|
'00:SCAN HDFS [tpch.lineitem, RANDOM]'
|
|
row_regex:.*partitions=1/1 files=1 size=.*
|
|
' runtime filters: RF000[bloom] -> l_orderkey'
|
|
' stored statistics:'
|
|
row_regex:.*table: rows=6001215 size=.*
|
|
' columns: all'
|
|
row_regex:.* extrapolated-rows=disabled max-scan-range-rows=[0-9]*.*
|
|
row_regex:.* mem-estimate=[0-9.]*[A-Z]*B mem-reservation=[0-9.]*[A-Z]*B thread-reservation=1.*
|
|
' tuple-ids=0 row-size=263B cardinality=6001215'
|
|
' in pipelines: 00(GETNEXT)'
|
|
====
|
|
---- QUERY
|
|
# Tests the warning about missing table stats in the explain header.
|
|
explain select count(t1.int_col), avg(t2.float_col), sum(t3.bigint_col)
|
|
from functional_avro.alltypes t1
|
|
inner join functional_parquet.alltypessmall t2 on (t1.id = t2.id)
|
|
left outer join functional_avro.alltypes t3 on (t2.id = t3.id)
|
|
where t1.month = 1 and t2.year = 2009 and t3.bool_col = false
|
|
---- RESULTS: VERIFY_IS_SUBSET
|
|
'Per-Host Resource Estimates: Memory=4.07GB'
|
|
'WARNING: The following tables are missing relevant table and/or column statistics.'
|
|
'functional_avro.alltypes, functional_parquet.alltypessmall'
|
|
====
|