mirror of
https://github.com/apache/impala.git
synced 2026-01-02 21:00:35 -05:00
9f678a7426
Reworks the FK/PK join detection logic to: - more accurately recognize many-to-many joins - avoid dim/dim joins for multi-column PKs The new detection logic maintains our existing philosophy of generally assuming a FK/PK join, unless there is strong evidence to the contrary, as follows. For each set of simple equi-join conjuncts between two tables, we compute the joint NDV of the right-hand side columns by multiplication, and if the joint NDV is significantly smaller than the right-hand side row count, then we are fairly confident that the right-hand side is not a PK. Otherwise, we assume the set of conjuncts could represent a FK/PK relationship. Extends the explain plan to include the outcome of the FK/PK detection at EXPLAIN_LEVEL > STANDARD. Performance testing: 1. Full TPC-DS run on 10TB: - Q10 improved by >100x - Q72 improved by >25x - Q17,Q26,Q29 improved by 2x - Q64 regressed by 10x - Total runtime: Improved by 2x - Geomean: Minor improvement The regression of Q64 is understood and we will try to address it in follow-on changes. The previous plan was better by accident and not because of superior logic. 2. Nightly TPC-H and TPC-DS runs: - No perf differences Testing: - The existing planner test cover the changes. - Code/hdfs run passed. Change-Id: I49074fe743a28573cff541ef7dbd0edd88892067 Reviewed-on: http://gerrit.cloudera.org:8080/7257 Reviewed-by: Alex Behm <alex.behm@cloudera.com> Tested-by: Impala Public Jenkins
61 lines
2.2 KiB
Plaintext
61 lines
2.2 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
|
|
'Per-Host Resource Reservation: Memory=136.00MB'
|
|
'Per-Host Resource Estimates: Memory=388.41MB'
|
|
''
|
|
'F02:PLAN FRAGMENT [UNPARTITIONED] hosts=1 instances=1'
|
|
'PLAN-ROOT SINK'
|
|
'| mem-estimate=0B mem-reservation=0B'
|
|
'|'
|
|
'04:EXCHANGE [UNPARTITIONED]'
|
|
'| mem-estimate=0B mem-reservation=0B'
|
|
'| tuple-ids=0,1 row-size=454B cardinality=5757710'
|
|
'|'
|
|
'F00:PLAN FRAGMENT [RANDOM] hosts=3 instances=3'
|
|
'02:HASH JOIN [INNER JOIN, BROADCAST]'
|
|
'| hash predicates: l_orderkey = o_orderkey'
|
|
'| fk/pk conjuncts: l_orderkey = o_orderkey'
|
|
'| runtime filters: RF000 <- o_orderkey'
|
|
'| mem-estimate=300.41MB mem-reservation=136.00MB'
|
|
'| tuple-ids=0,1 row-size=454B cardinality=5757710'
|
|
'|'
|
|
'|--03:EXCHANGE [BROADCAST]'
|
|
'| | mem-estimate=0B mem-reservation=0B'
|
|
'| | tuple-ids=1 row-size=191B cardinality=1500000'
|
|
'| |'
|
|
'| F01:PLAN FRAGMENT [RANDOM] hosts=2 instances=2'
|
|
'| 01:SCAN HDFS [tpch.orders, RANDOM]'
|
|
row_regex:.*partitions=1/1 files=1 size=.*
|
|
'| stats-rows=1500000 extrapolated-rows=disabled'
|
|
'| table stats: rows=1500000 size=162.56MB'
|
|
'| column stats: all'
|
|
'| mem-estimate=88.00MB mem-reservation=0B'
|
|
'| tuple-ids=1 row-size=191B cardinality=1500000'
|
|
'|'
|
|
'00:SCAN HDFS [tpch.lineitem, RANDOM]'
|
|
row_regex:.*partitions=1/1 files=1 size=.*
|
|
' runtime filters: RF000 -> l_orderkey'
|
|
' stats-rows=6001215 extrapolated-rows=disabled'
|
|
' table stats: rows=6001215 size=718.94MB'
|
|
' column stats: all'
|
|
' mem-estimate=88.00MB mem-reservation=0B'
|
|
' tuple-ids=0 row-size=263B cardinality=6001215'
|
|
====
|
|
---- 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=4.03GB'
|
|
'WARNING: The following tables are missing relevant table and/or column statistics.'
|
|
'functional_avro.alltypes, functional_parquet.alltypessmall'
|
|
====
|