Question

i am running into a problem, i have huge database and now i cant make changes i need to process all this data first. Problem is that each query which is writen below takes about 1-2 Minutes. If anyone know how to make a optimized select i would apprichiate it.

select distinct x.peerId
,(select max(id) from checkPeer where peerId=x.peerId) as idd 
from checkPeer x where x.id>0 order by x.id desc

Is there way to optimize this select from below defined table?

CREATE TABLE `checkPeer` (
`id` bigint(20) NOT NULL AUTO_INCREMENT,
`peerId` varchar(100) COLLATE utf8_bin DEFAULT NULL,
`peerDate` datetime DEFAULT NULL,
`peerUrl` bigint(20) DEFAULT '0',
`peerState` bigint(20) DEFAULT '0',
`peerMessage` varchar(500) COLLATE utf8_bin DEFAULT NULL,
`peerResource` varchar(5000) COLLATE utf8_bin DEFAULT NULL,
PRIMARY KEY (`id`),
KEY `peerId` (`peerId`),
KEY `peerUrl` (`peerUrl`)
) ENGINE=MyISAM AUTO_INCREMENT=688791 DEFAULT CHARSET=utf8 COLLATE=utf8_bin;

Thank you!

Answer 1

a few immediate suggestions, remove DISTINCT and try the following:

select x.peerId, x1.id as idd 
from checkPeer x 
left join
(
    select max(id) id, peerid
    from checkPeer
    group by peerid
) x1
    on x.peerid = x1.peerid
where x.id>0 
order by x.id desc

The above works great, if you wind up needing more fields but you stated you only need the two so you can try:

select peerId, max(id) as idd
from checkPeer 
where id > 0
group by peerid