Mysql BIgQuery一对多关系

Question

I had create tables with one to many relationship , size of rows 11328441 . The table structure is

CREATE TABLE `gameresnum` (
  `ID` bigint(20) NOT NULL,
  `DRAWDAY` date DEFAULT NULL,
  `DRAWNO` int(11) DEFAULT NULL,
  `DRAWTIME` time DEFAULT NULL,
  `RESNUM` int(11) NOT NULL
) ENGINE=InnoDB DEFAULT CHARSET=latin1;

ALTER TABLE `gameresnum`
  ADD PRIMARY KEY (`ID`),
  ADD KEY `RESNUM` (`RESNUM`);

.

I tried to make this query :

SELECT count(*) as count FROM gameresnum gone
JOIN gameresnum gtwo ON gone.DRAWNO=gtwo.DRAWNO WHERE gone.RESNUM=1 AND gtwo.RESNUM=2

But it is very slow up to 2 mint.

Answer 1

your drawno column does not have an index. thats why.

Answer 2

Your query would benefit from

INDEX(resnum, drawno) -- composite and covering

You could also experiment with

SELECT COUNT(*)
    FROM (
        SELECT 1
            FROM gamesresnum
            WHERE resnum IN (1, 2)
            GROUP BY drawno
            HAVING COUNT(*) = 2  -- to see if both are present
         ) AS x

It needs INDEX(drawno, resnum) .

Caveat: This reformulation assumes that (drawno, resnum) is unique. I don't know if this will be faster.