Speeding up my queries in PHP - php

Question

I'm working on trying to speed up a webpage I have created. I know the issue is that I have a query within a query. I feel like there has to be a quicker way to accomplish the same results, but I'm running out of ideas. (My first attempt at this took 45 seconds for the page to load, now I'm down to about 6)
What I'm trying to do is pull run rate information from tables. I need to pull the correct startup and end of run rates from the runrate table, but all I have to go off of initially is the workcenter ID.
I feel like if the tables were set up a little bit better then it probably would've have been so difficult, but it's what I inherited and as a result I'm a bit stuck. I need to pull a month worth of data from each workcenter (about 15) where there can be as many as 4-5 runs each day... Quite a bit of data to process.
Here's the PHP code:
$qtotalStartup = mysql_query("
SELECT startup.recordID, startup.date, startup.time, runrate.rate AS temRate, runrate.formID
FROM jos_a_inproc_startup startup JOIN jos_a_runrate runrate ON startup.recordID = runrate.recordID
WHERE startup.workcenterId = $id AND runrate.rate > 0 AND runrate.formID = 1 AND startup.date > DATE_SUB(NOW(), INTERVAL 1 MONTH)") or die(mysql_error());
$totalStartCtr = mysql_num_rows($qtotalStartup);
if ($totalStartCtr > 0) {
while($rtotalStartup = mysql_fetch_assoc($qtotalStartup)) {
$hours = 0;
$goalRate = 0;
$sumHrRR = 0;
$startDate = 0;
$startTime = 0;
$startupNum = $rtotalStartup['recordID'];
$goalRate = $rtotalStartup['temRate'];
$startDate = $rtotalStartup['date'];
$startTime = $rtotalStartup['time'];
$startTime = strtotime($startDate . ' ' . $startTime);
//now that we have all of the startup form info, we can move to the end of run information
//this query will retrieve the correct date, time, and ending run rate for us to use with our calculations.
$qtotalEOR = mysql_query("
SELECT eor.recordID AS eorRec, eor.date, eor.time, eor.startupid, runrate1.rate AS tempRate, runrate1.formID
FROM jos_a_inproc_eor eor JOIN jos_a_runrate runrate1 ON eor.recordID = runrate1.recordID
WHERE eor.startupid = $startupNum AND runrate1.rate > 0 AND runrate1.formID = 3") or die(mysql_error());
$totalEORCtr = mysql_num_rows($qtotalEOR);
if ($totalEORCtr > 0) {
while($rtotalEOR = mysql_fetch_assoc($qtotalEOR)) {
//reset the accumulator to 0 so we don't get extra 'bad' data.
$sumHrRR = 0;
$newGoalRate = 0;
$lastestDate = 0;
$latestTime = 0;
$eorNum = $rtotalEOR['eorRec'];
$latestDate = $rtotalEOR['date'];
$latestTime = $rtotalEOR['time'];
$latestTime = strtotime($latestDate . ' ' . $latestTime);
$sumHrRR= $rtotalEOR['tempRate'];
Any ideas would be greatly appreciated. I know it may be difficult to understand what I'm trying to get at without much more information, so let me know if you need to know anything else. Thanks.

Answer 1

Maby try using multiple JOINS like this one:
SELECT startup.recordID, startup.date, startup.time,
runrate.rate AS temRate, runrate.formID
-- stuff from second query
eor.recordID AS eorRec, eor.date AS eor_date,
eor.time AS eor_time, eor.startupid AS eor_startupid,
runrate1.rate AS eor_tempRate,
runrate1.formID AS runrate1_formID
FROM jos_a_inproc_startup startup
JOIN jos_a_runrate runrate ON startup.recordID = runrate.recordID
-- second query LEFT JOIN
LEFT JOIN jos_a_inproc_eor eor
ON eor.startupid = startup.recordID
LEFT JOIN jos_a_runrate runrate1
ON eor.recordID = runrate1.recordID
AND runrate1.rate > 0
AND runrate1.formID = 3
WHERE startup.workcenterId = $id
AND runrate.rate > 0
AND runrate.formID = 1
AND startup.date > DATE_SUB(NOW(), INTERVAL 1 MONTH)
I don't know if I'm right but I think that you are also doing some aggregation work with results inside PHP. You could do it inside database using like sum() or avg() and GROUP BY. You will save some time when transfering smaller result set from database to server and time for looping and aggregating inside PHP. Also most of the time using JOIN is much faster than using queries in loop or even subqueries inside query.
You should also check if indexes are set on columns you search in. Also use EXPLAIN to check how query is executed.

Answer 2

you can use Mem-Cache techniques to make it much faster ,and try to make your queries the simpler that u can .. dont retrieve values that you dont use in your scripts ..

Answer 3

How many records are you typically dealing with as output? How big are the tables? Have you reviewed the indexes? Have you analyzed them recently (rebuilt them)?
Also, are you sending the data back to the browser using deflate? See:
http://httpd.apache.org/docs/2.2/mod/mod_deflate.html

Answer 4

Well, you could try using multiple INNER JOINs (see) and have only one query instead of one query inside a query, which greatly impacts on performance. You could try something like this, and tweaking it a little:
SELECT
startup.recordID AS startupRecordID,
startup.date AS startupDate,
startup.time AS startupTime,
runrate.rate,
runrate.formID,
eor.recordID AS eorRecordID,
eor.date AS eorDate,
eor.time AS eorTime,
eor.startupid AS eorStartupID
FROM jos_a_inproc_startup startup
INNER JOIN jos_a_runrate runrate
ON startup.recordID = runrate.recordID
INNER JOIN jos_a_inproc_eor eor
ON startup.recordID = eor.startupid
WHERE
startup.workcenterId = $id
AND runrate.rate > 0
AND runrate.formID = 1
AND startup.date > DATE_SUB(NOW(), INTERVAL 1 MONTH)