mirror of
https://github.com/apache/impala.git
synced 2026-02-02 06:00:36 -05:00
9549d3859e
The bug was that filter routing table construction removed filters from the TPlanNode structure for the join when a finstance was not a producer of that filter. The TPlanNode is shared between all instances of a fragment on a backend, so this meant that the filter was removed for all instances on that backend, often meaning that no filters would be produced at all. It was awkward fixing the bug within the framework of the current data structures, where the routing table is keyed by filter_id, so I ended up refactoring the routing table somewhat. This also allowed fixing a TODO about O(n^2) construction of the routing table. Testing: Add regression test that timed out without fix. Perf: Ran a single node TPC-H workload with scale factor 30. No perf change. Change-Id: I26e3628a982d5d9b8b24eb96b28aff11f8aa6669 Reviewed-on: http://gerrit.cloudera.org:8080/14511 Reviewed-by: Impala Public Jenkins <impala-public-jenkins@cloudera.com> Tested-by: Impala Public Jenkins <impala-public-jenkins@cloudera.com>
52 lines
1.7 KiB
Plaintext
52 lines
1.7 KiB
Plaintext
====
|
|
---- QUERY
|
|
# Test that the right amount of memory is reserved for this plan - the planner assumes
|
|
# that the join build does not run in parallel with the left side of the join.
|
|
set num_nodes=1;
|
|
set runtime_filter_wait_time_ms=$RUNTIME_FILTER_WAIT_TIME_MS;
|
|
set runtime_filter_mode=LOCAL;
|
|
select STRAIGHT_JOIN count(*) from alltypes p join [BROADCAST] alltypestiny b
|
|
on p.month = b.int_col and b.month = 1 and b.string_col = "1"
|
|
---- RESULTS
|
|
620
|
|
---- RUNTIME_PROFILE
|
|
row_regex: .*Files rejected: 22 \(22\).*
|
|
====
|
|
---- QUERY
|
|
# Test that the single-node plan is executable and produces the correct result for a
|
|
# nested loop join mode that can't be executed in a distributed fashion.
|
|
set num_nodes=1;
|
|
select straight_join count(*)
|
|
from functional.alltypestiny a right outer join functional.alltypes b
|
|
on a.id != b.id or a.int_col < b.int_col
|
|
right outer join functional.alltypesagg c
|
|
on a.smallint_col >= c.smallint_col
|
|
where a.id < 10 and c.bigint_col = 10
|
|
---- RESULTS
|
|
291990
|
|
====
|
|
---- QUERY
|
|
# Regression IMPALA-9019: fix runtime filter propagation with mt_dop.
|
|
# This plan has a global BROADCAST filter sent from the scan of alltypestiny
|
|
# to the scan of alltypes. Before the bug was fixed, this filter would not
|
|
# be produced, but was expected by the destination. Setting the wait time
|
|
# higher than the time limit caused this query to reliably fail because
|
|
# the destination scan would block waiting for the filter to arrive.
|
|
set exec_time_limit_s=30;
|
|
set runtime_filter_wait_time_ms=120000;
|
|
select straight_join t1.id, t1.int_col
|
|
from (select distinct * from alltypes) t1
|
|
join /*+broadcast*/ alltypestiny t2 on t1.id = t2.id;
|
|
---- TYPES
|
|
int,int
|
|
---- RESULTS
|
|
0,0
|
|
1,1
|
|
2,2
|
|
3,3
|
|
4,4
|
|
5,5
|
|
6,6
|
|
7,7
|
|
====
|