1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
|
#
# index_merge tests for statements using intersect algorithm
#
--disable_warnings
DROP TABLE IF EXISTS t1,t2;
--enable_warnings
CREATE TABLE t1 (
pk MEDIUMINT NOT NULL AUTO_INCREMENT,
a INT NOT NULL,
b INT NOT NULL,
c INT NOT NULL,
d INT NOT NULL,
PRIMARY KEY pk (pk),
KEY idx_a (a),
KEY idx_b (b),
KEY idx_c (c),
KEY idx_d (d)
);
--disable_query_log
# Inserting a lot of rows inorder to enable index_merge intersect
INSERT INTO t1(a,b,c,d) VALUES
( RAND()*5, RAND()*5, RAND()*5, RAND()*5 );
let $cnt=4;
while ($cnt)
{
INSERT INTO t1(a,b,c,d) SELECT 6,6,6,6 FROM t1;
dec $cnt;
}
INSERT INTO t1(a,b,c,d) SELECT 6, RAND()*5, RAND()*5,
RAND()*5 FROM t1 LIMIT 3;
INSERT INTO t1(a,b,c,d) SELECT RAND()*5, 6, RAND()*5,
RAND()*5 FROM t1 LIMIT 3;
INSERT INTO t1(a,b,c,d) SELECT RAND()*5, RAND()*5, 6,
RAND()*5 FROM t1 LIMIT 3;
INSERT INTO t1(a,b,c,d) SELECT RAND()*5, RAND()*5,
RAND()*5, 6 FROM t1 LIMIT 3;
let $cnt=7;
while ($cnt)
{
INSERT INTO t1(a,b,c,d) SELECT RAND()*5, RAND()*5,
RAND()*5, RAND()*5 FROM t1;
dec $cnt;
}
--enable_query_log
# The following statement analyzes and
# stores the key distribution for a table.
ANALYZE TABLE t1;
# DELETEs are not included as index merge intersection
# is disabled for DELETE statements.
# 1. UPDATE
--let $query = WHERE b=6 AND c=6
--replace_result "idx_c,idx_b" "idx_b,idx_c"
--replace_column 9 #
--eval EXPLAIN UPDATE t1 SET a=2 $query
--eval SELECT COUNT(*), SUM(a) FROM t1 $query
--eval UPDATE t1 SET a=2 $query
--eval SELECT COUNT(*), SUM(a) FROM t1 $query
--let $query = WHERE b=6 AND c=6 AND d=6
--replace_result idx_b idx_x idx_c idx_x idx_d idx_x
--replace_column 9 #
--eval EXPLAIN UPDATE t1 SET a=2 $query
--eval SELECT COUNT(*), SUM(a) FROM t1 $query
--eval UPDATE t1 SET a=2 $query
--eval SELECT COUNT(*), SUM(a) FROM t1 $query
--let $query = WHERE d=6 AND a IS NOT NULL AND b=6
--replace_result "idx_d,idx_b" "idx_b,idx_d"
--replace_column 9 #
--eval EXPLAIN UPDATE t1 SET c=6 $query
--eval SELECT COUNT(*), SUM(c) FROM t1 $query
--eval UPDATE t1 SET c=6 $query
--eval SELECT COUNT(*), SUM(c) FROM t1 $query
--let $query = WHERE d=6 AND a=6 AND c <> 6
--replace_column 9 #
--eval EXPLAIN UPDATE t1 SET b=0 $query
--eval SELECT COUNT(*), SUM(b) FROM t1 $query
--eval UPDATE t1 SET b=0 $query
--eval SELECT COUNT(*), SUM(b) FROM t1 $query
--let $query = WHERE d=6 AND a=6 AND c IN (1,2,3,4,5)
--replace_column 9 #
--eval EXPLAIN UPDATE t1 SET a=100 $query
--eval SELECT COUNT(*), SUM(a) FROM t1 $query
--eval UPDATE t1 SET a=100 $query
--eval SELECT COUNT(*), SUM(a) FROM t1 $query
# uses range scan instead of index_merge
--let $query = WHERE a=5 AND b=4 AND d<3
--replace_column 9 #
--eval EXPLAIN UPDATE t1 SET a=2 $query
--eval UPDATE t1 SET a=2 $query
# Any range condition over a primary key of an InnoDB table.
create table t2 (
pk MEDIUMINT NOT NULL AUTO_INCREMENT,
a INT NOT NULL,
b INT NOT NULL,
c INT NOT NULL,
d INT NOT NULL,
PRIMARY KEY pk (pk),
KEY idx_a (a),
KEY idx_b (b),
KEY idx_c (c),
KEY idx_d (d)
)engine=InnoDB;
INSERT INTO t2 SELECT * FROM t1;
ANALYZE TABLE t2;
--let $query = WHERE pk<2492 AND d=1
--replace_column 9 #
--eval EXPLAIN UPDATE t2 SET a=2 $query
--eval UPDATE t2 SET a=2 $query
# 2. REPLACE and INSERT
CREATE TABLE t3(
a INT NOT NULL,
b INT NOT NULL,
c INT NOT NULL,
d INT NOT NULL,
e INT
);
INSERT INTO t3(a,b,c,d,e) VALUES (3890,3890,3890,3890,3890);
INSERT INTO t3(a,b,c,d,e) VALUES (4000,4000,4000,4000,4000);
--let $query = INTO t3 SELECT * FROM t1 WHERE b=6 AND c=6
--replace_result "idx_c,idx_b" "idx_b,idx_c"
--replace_column 9 #
--eval EXPLAIN REPLACE $query
--eval SELECT COUNT(*) FROM t3
--eval REPLACE $query
--eval SELECT COUNT(*) FROM t3
--let $query = INTO t3 SELECT * FROM t1 WHERE d=6 AND a IS NOT NULL AND b=6
--replace_column 9 #
--replace_result "idx_d,idx_b" "idx_b,idx_d"
--eval EXPLAIN INSERT $query
--eval SELECT COUNT(*) FROM t3
--eval INSERT $query
--eval SELECT COUNT(*) FROM t3
# Test case for multi column set-up.
CREATE TABLE t4 (
pk MEDIUMINT NOT NULL AUTO_INCREMENT,
a INT NOT NULL,
b INT NOT NULL,
c INT NOT NULL,
d INT NOT NULL,
PRIMARY KEY pk (pk),
KEY idx_cd (c,d),
KEY idx_bd (b,d)
);
INSERT INTO t4 SELECT * FROM t1;
# The following statement analyzes and
# stores the key distribution for a table.
ANALYZE TABLE t4;
--let $query = WHERE b=6 AND c=6 AND d=6
--replace_result idx_b idx_x idx_c idx_x idx_d idx_x
--replace_column 9 #
--eval EXPLAIN UPDATE t4 SET a=2 $query
--eval SELECT COUNT(*), SUM(a) FROM t4 $query
--eval UPDATE t4 SET a=2 $query
--eval SELECT COUNT(*), SUM(a) FROM t4 $query
DROP TABLE t1,t2,t3,t4;
#
# end of test cases for intersect index_merge optimization technique
#
|
