Slow queries with MySQL large database - php

Question

I have a table ipaddresses that contains IP ranges. Columns are: ipStart, ipEnd
Example of
ipStart: 3579374832
ipEnd: 3579374839
Now I want to select 300ish IP addresses from another table: visit_count. And based on those 300 addresses I want to check if the ip address is in the range of any of IPs in the table ipaddresses
My current code:
$results = $database->query("SELECT user_id,
DATE_FORMAT(last_visit, '%Y-%m-%d') as datumet,
cookieId, ipaddress
FROM `visit_count`
WHERE last_visit BETWEEN
'$firstDayOfMonth' AND '$lastDayOfMonth'
AND user_id = '$userId'
GROUP BY cookieId
ORDER BY last_visit ASC");
$rows = $database->loadObjectList($results);
foreach ( $rows as $row ) {
$ipaddressLong = ip2long($row->ipaddress);
// This is where its very very slow
$selO = $database->query("SELECT ipStart, ipEnd FROM `ipaddresses`
WHERE '$ipaddressLong' BETWEEN `ipStart` AND `ipEnd`");
$rowO = $database->getrow($selO);
}
The result is a really slow query that consumes a lot of cpu.
ipaddresses has indexes for both ipStart and ipEnd and contains around 50k rows.
How can I make this go faster?

Answer 1

A quick and dirty change to a single query using a JOIN:-
SELECT a.ipStart, a.ipEnd
FROM `ipaddresses` a
INNER JOIN
(
SELECT user_id, DATE_FORMAT(last_visit, '%Y-%m-%d') as datumet, cookieId, inet_aton(ipaddress ) AS ipaddressLong
FROM `visit_count`
WHERE last_visit BETWEEN '$firstDayOfMonth' AND '$lastDayOfMonth'
AND user_id = '$userId'
GROUP BY cookieId
) b
ON b.ipaddressLong BETWEEN a.ipStart AND a.ipEnd
Note that this could probably be simplified. However it depends on your use of GROUP BY cookieId . At the moment this is going to get one row per cookie id, but which row is undefined (ie, could be any user_id, ip address and last visit date that goes with that cookie id).

Answer 2

You can simplify the first select. You don't need to select any fields beyond ipaddress. You also don't need to order or group the result. If you want unique ipaddresses, just use distinct
select distinct ipaddress
from visit_count
where last_visit between '$firstDayOfMonth' and '$lastDayOfMonth'
and user_id = '$userId'
300 selects don't seem to be much, even with the second table having 50k rows. On my machine, even without indexes, it takes only a fraction of a second.
If you want to simplify it anyway, you can join the two tables with
select ipstart, ipend
from ipaddresses i
join visit_count v on inet_aton(v.ipaddress) between i.ipstart and i.ipend
where v.last_visit between '$firstDayOfMonth' and '$lastDayOfMonth'
and v.userid = '$userId'

Answer 3

1) Use better query like other suggested to you
2) Have you defined indexes for the field you need ?
3) Dont know if you use mysql or mysqli...anyway use fetch_assoc that is faster then fetch_array
with this 3 trick you should be ok