mirror of
https://github.com/MariaDB/server.git
synced 2025-01-29 02:05:57 +01:00
3fa99f0c0e
The main difference in code path between EQ_REF and REF is that for REF we have to do an extra read_next on the index to check that there is no more matching rows. Before this patch we added a preference of EQ_REF by ensuring that REF would always estimate to find at least 2 rows. This patch adds the cost of the extra key read_next to REF access and removes the code that limited REF to at least 2 rows. For some queries this can have a big effect as the total estimated rows will be halved for each REF table with 1 rows. multi_range cost calculations are also changed to take into account the difference between EQ_REF and REF. The effect of the patch to the test suite: - About 80 test case changed - Almost all changes where for EXPLAIN where estimated rows for REF where changed from 2 to 1. - A few test cases using explain extended had a change of 'filtered'. This is because of the estimated rows are now closer to the calculated selectivity. - A very few test had a change of table order. This is because the change of estimated rows from 2 to 1 or the small cost change for REF (main.subselect_sj_jcl6, main.group_by, main.dervied_cond_pushdown, main.distinct, main.join_nested, main.order_by, main.join_cache) - No key statistics and the estimated rows are now smaller which cased estimated filtering to be lower. (main.subselect_sj_mat) - The number of total rows are halved. (main.derived_cond_pushdown) - Plans with 1 row changed to use RANGE instead of REF. (main.group_min_max) - ALL changed to REF (main.key_diff) - Key changed from ref + index_only to PRIMARY key for InnoDB, as OPTIMIZER_ROW_LOOKUP_COST + OPTIMIZER_ROW_NEXT_FIND_COST is smaller than OPTIMIZER_KEY_LOOKUP_COST + OPTIMIZER_KEY_NEXT_FIND_COST. (main.join_outer_innodb) - Cost changes printouts (main.opt_trace*) - Result order change (innodb_gis.rtree)
369 lines
9.7 KiB
Text
369 lines
9.7 KiB
Text
create or replace table t1 (a int, b int, c int, key(a,c), key(b,c), key (c,b)) engine=aria;
|
|
insert into t1 select seq/100+1, mod(seq,10), mod(seq,15) from seq_1_to_10000;
|
|
insert into t1 select seq/100+1, mod(seq,10), 10 from seq_1_to_1000;
|
|
optimize table t1;
|
|
Table Op Msg_type Msg_text
|
|
test.t1 optimize status OK
|
|
select count(*) from t1 where a=2;
|
|
count(*)
|
|
200
|
|
select count(*) from t1 where b=5;
|
|
count(*)
|
|
1100
|
|
select count(*) from t1 where c=5;
|
|
count(*)
|
|
667
|
|
select count(*) from t1 where c=10;
|
|
count(*)
|
|
1667
|
|
select count(*) from t1 where a=2 and b=5;
|
|
count(*)
|
|
20
|
|
select count(*) from t1 where c=10 and b=5;
|
|
count(*)
|
|
433
|
|
select count(*) from t1 where c=5 and b=5;
|
|
count(*)
|
|
334
|
|
set optimizer_trace="enabled=on";
|
|
select count(*) from t1 where a=2 and b=5 and c=10;
|
|
count(*)
|
|
14
|
|
set @trace=(select trace from INFORMATION_SCHEMA.OPTIMIZER_TRACE);
|
|
select
|
|
JSON_DETAILED(
|
|
JSON_EXTRACT(
|
|
JSON_EXTRACT(@trace, '$**.considered_execution_plans'),
|
|
'$[0]'
|
|
)
|
|
) as JS;
|
|
JS
|
|
[
|
|
{
|
|
"plan_prefix": "",
|
|
"get_costs_for_tables":
|
|
[
|
|
{
|
|
"best_access_path":
|
|
{
|
|
"table": "t1",
|
|
"plan_details":
|
|
{
|
|
"record_count": 1
|
|
},
|
|
"considered_access_paths":
|
|
[
|
|
{
|
|
"access_type": "ref",
|
|
"index": "a",
|
|
"used_range_estimates": true,
|
|
"rows": 104,
|
|
"cost": 0.060988785,
|
|
"chosen": true
|
|
},
|
|
{
|
|
"access_type": "ref",
|
|
"index": "b",
|
|
"used_range_estimates": true,
|
|
"rows": 340,
|
|
"cost": 0.141618657,
|
|
"chosen": false,
|
|
"cause": "cost"
|
|
},
|
|
{
|
|
"access_type": "ref",
|
|
"index": "c",
|
|
"used_range_estimates": true,
|
|
"rows": 632,
|
|
"cost": 0.241826241,
|
|
"chosen": false,
|
|
"cause": "cost"
|
|
},
|
|
{
|
|
"access_type": "index_merge",
|
|
"rows": 7,
|
|
"rows_after_filter": 7,
|
|
"rows_out": 7,
|
|
"cost": 0.045367017,
|
|
"chosen": true
|
|
}
|
|
],
|
|
"chosen_access_method":
|
|
{
|
|
"type": "index_merge",
|
|
"rows_read": 7,
|
|
"rows_out": 7,
|
|
"cost": 0.045367017,
|
|
"uses_join_buffering": false
|
|
}
|
|
}
|
|
}
|
|
]
|
|
},
|
|
{
|
|
"plan_prefix": "",
|
|
"table": "t1",
|
|
"rows_for_plan": 7,
|
|
"cost_for_plan": 0.045367017
|
|
}
|
|
]
|
|
select JSON_DETAILED(JSON_EXTRACT(@trace, '$**.selectivity_for_indexes')) as JS;
|
|
JS
|
|
[
|
|
[
|
|
{
|
|
"index_name": "a",
|
|
"selectivity_from_index": 0.009454545
|
|
},
|
|
{
|
|
"index_name": "b",
|
|
"selectivity_from_index": 0.1
|
|
},
|
|
{
|
|
"use_opt_range_condition_rows_selectivity": 6.363636e-4
|
|
}
|
|
]
|
|
]
|
|
select count(*) from t1 where a=2 and b=5 and c=5;
|
|
count(*)
|
|
3
|
|
set @trace=(select trace from INFORMATION_SCHEMA.OPTIMIZER_TRACE);
|
|
select
|
|
JSON_DETAILED(
|
|
JSON_EXTRACT(
|
|
JSON_EXTRACT(@trace, '$**.considered_execution_plans'),
|
|
'$[0]'
|
|
)
|
|
) as JS;
|
|
JS
|
|
[
|
|
{
|
|
"plan_prefix": "",
|
|
"get_costs_for_tables":
|
|
[
|
|
{
|
|
"best_access_path":
|
|
{
|
|
"table": "t1",
|
|
"plan_details":
|
|
{
|
|
"record_count": 1
|
|
},
|
|
"considered_access_paths":
|
|
[
|
|
{
|
|
"access_type": "ref",
|
|
"index": "a",
|
|
"used_range_estimates": true,
|
|
"rows": 6,
|
|
"cost": 0.005388489,
|
|
"chosen": true
|
|
},
|
|
{
|
|
"access_type": "ref",
|
|
"index": "b",
|
|
"used_range_estimates": true,
|
|
"rows": 232,
|
|
"cost": 0.104720241,
|
|
"chosen": false,
|
|
"cause": "cost"
|
|
},
|
|
{
|
|
"access_type": "ref",
|
|
"index": "c",
|
|
"used_range_estimates": true,
|
|
"rows": 293,
|
|
"cost": 0.125561013,
|
|
"chosen": false,
|
|
"cause": "cost"
|
|
},
|
|
{
|
|
"type": "scan",
|
|
"chosen": false,
|
|
"cause": "cost"
|
|
}
|
|
],
|
|
"chosen_access_method":
|
|
{
|
|
"type": "ref",
|
|
"rows_read": 6,
|
|
"rows_out": 0.6,
|
|
"cost": 0.005388489,
|
|
"uses_join_buffering": false
|
|
}
|
|
}
|
|
}
|
|
]
|
|
},
|
|
{
|
|
"plan_prefix": "",
|
|
"table": "t1",
|
|
"rows_for_plan": 0.6,
|
|
"cost_for_plan": 0.005388489,
|
|
"pushdown_cond_selectivity": 0.1,
|
|
"filtered": 10,
|
|
"rows_out": 0.6
|
|
}
|
|
]
|
|
select JSON_DETAILED(JSON_EXTRACT(@trace, '$**.selectivity_for_indexes')) as JS;
|
|
JS
|
|
[
|
|
[
|
|
{
|
|
"index_name": "a",
|
|
"selectivity_from_index": 5.454545e-4
|
|
},
|
|
{
|
|
"index_name": "b",
|
|
"selectivity_from_index": 0.1
|
|
}
|
|
]
|
|
]
|
|
# Ensure that we only use selectivity from non used index for simple cases
|
|
select count(*) from t1 where (a=2 and b= 5);
|
|
count(*)
|
|
20
|
|
select JSON_DETAILED(JSON_EXTRACT(trace, '$**.selectivity_for_indexes')) as JS
|
|
from INFORMATION_SCHEMA.OPTIMIZER_TRACE;
|
|
JS
|
|
[
|
|
[
|
|
{
|
|
"index_name": "a",
|
|
"selectivity_from_index": 0.017545455
|
|
},
|
|
{
|
|
"index_name": "b",
|
|
"selectivity_from_index": 0.073181818
|
|
}
|
|
]
|
|
]
|
|
# All of the following should have selectivity=1 for index 'b'
|
|
select count(*) from t1 where (a=2 and b between 0 and 100);
|
|
count(*)
|
|
200
|
|
select JSON_DETAILED(JSON_EXTRACT(trace, '$**.selectivity_for_indexes')) as JS
|
|
from INFORMATION_SCHEMA.OPTIMIZER_TRACE;
|
|
JS
|
|
[
|
|
[
|
|
{
|
|
"index_name": "a",
|
|
"selectivity_from_index": 0.017545455
|
|
},
|
|
{
|
|
"index_name": "b",
|
|
"selectivity_from_index": 1
|
|
}
|
|
]
|
|
]
|
|
select count(*) from t1 where (a in (2,3) and b between 0 and 100);
|
|
count(*)
|
|
400
|
|
select JSON_DETAILED(JSON_EXTRACT(trace, '$**.selectivity_for_indexes')) as JS
|
|
from INFORMATION_SCHEMA.OPTIMIZER_TRACE;
|
|
JS
|
|
[
|
|
[
|
|
{
|
|
"index_name": "a",
|
|
"selectivity_from_index": 0.035090909
|
|
},
|
|
{
|
|
"index_name": "b",
|
|
"selectivity_from_index": 1
|
|
}
|
|
]
|
|
]
|
|
select count(*) from t1 where (a>2 and b between 0 and 100);
|
|
count(*)
|
|
10702
|
|
select JSON_DETAILED(JSON_EXTRACT(trace, '$**.selectivity_for_indexes')) as JS
|
|
from INFORMATION_SCHEMA.OPTIMIZER_TRACE;
|
|
JS
|
|
[
|
|
[
|
|
{
|
|
"index_name": "a",
|
|
"selectivity_from_index": 0.973909091
|
|
},
|
|
{
|
|
"index_name": "b",
|
|
"selectivity_from_index": 1
|
|
}
|
|
]
|
|
]
|
|
select count(*) from t1 where (a>=2 and b between 0 and 100);
|
|
count(*)
|
|
10902
|
|
select JSON_DETAILED(JSON_EXTRACT(trace, '$**.selectivity_for_indexes')) as JS
|
|
from INFORMATION_SCHEMA.OPTIMIZER_TRACE;
|
|
JS
|
|
[
|
|
[
|
|
{
|
|
"index_name": "a",
|
|
"selectivity_from_index": 0.991454545
|
|
},
|
|
{
|
|
"index_name": "b",
|
|
"selectivity_from_index": 1
|
|
}
|
|
]
|
|
]
|
|
select count(*) from t1 where (a<=2 and b between 0 and 100);
|
|
count(*)
|
|
298
|
|
select JSON_DETAILED(JSON_EXTRACT(trace, '$**.selectivity_for_indexes')) as JS
|
|
from INFORMATION_SCHEMA.OPTIMIZER_TRACE;
|
|
JS
|
|
[
|
|
[
|
|
{
|
|
"index_name": "a",
|
|
"selectivity_from_index": 0.026181818
|
|
},
|
|
{
|
|
"index_name": "b",
|
|
"selectivity_from_index": 1
|
|
}
|
|
]
|
|
]
|
|
select count(*) from t1 where (a<2 and b between 0 and 100);
|
|
count(*)
|
|
98
|
|
select JSON_DETAILED(JSON_EXTRACT(trace, '$**.selectivity_for_indexes')) as JS
|
|
from INFORMATION_SCHEMA.OPTIMIZER_TRACE;
|
|
JS
|
|
[
|
|
[
|
|
{
|
|
"index_name": "a",
|
|
"selectivity_from_index": 0.008636364
|
|
},
|
|
{
|
|
"index_name": "b",
|
|
"selectivity_from_index": 1
|
|
}
|
|
]
|
|
]
|
|
select count(*) from t1 where (a between 2 and 3 and b between 0 and 100);
|
|
count(*)
|
|
400
|
|
select JSON_DETAILED(JSON_EXTRACT(trace, '$**.selectivity_for_indexes')) as JS
|
|
from INFORMATION_SCHEMA.OPTIMIZER_TRACE;
|
|
JS
|
|
[
|
|
[
|
|
{
|
|
"index_name": "a",
|
|
"selectivity_from_index": 0.035090909
|
|
},
|
|
{
|
|
"index_name": "b",
|
|
"selectivity_from_index": 1
|
|
}
|
|
]
|
|
]
|
|
drop table t1;
|
|
set optimizer_trace='enabled=off';
|