MYSQL Inner Join speed issue - php

Question

I have been having mega issues with this query.
It's a file that's run by cron every 24 hours and runs conditional checks against all members in the database.
Apparently it crashes the MYSQL server (takes 2 hours to execute?).
Every table has primary keys set on rows foo_id & foo_uid so the joins are indexed under PRIMARY and as such should be good for speed.
please help, this is killing me.
$members = new WA_MySQLi_RS("members", $alpha, 1);
$members->setQuery("SELECT
registration.*,
child_base_survey.*,
child_base_scas.*,
child_base_smqf.*,
parent_base_survey.*,
parent_base_ippa.*,
parent_base_eac.*,
parent_base_scas.*,
parent_base_smqf.*,
parent_base_eval.*,
user_access_level.*,
parent_one_month_survey.*,
parent_one_month_ippa.*,
parent_one_month_eac.*,
parent_one_month_eval.*,
child_three_month_survey.*,
child_three_month_scas.*,
child_three_month_smqf.*,
parent_three_month_survey.*,
parent_three_month_scas.*,
parent_three_month_smqf.*,
parent_three_month_eval.*,
cron.*
FROM registration
INNER JOIN child_base_survey ON registration.rego_parent_uid = child_base_survey.child_base_survey_uid
INNER JOIN child_base_scas ON child_base_survey.child_base_survey_uid = child_base_scas.child_base_scas_uid
INNER JOIN child_base_smqf ON child_base_scas.child_base_scas_uid = child_base_smqf.child_base_smqf_uid
INNER JOIN parent_base_survey ON child_base_smqf.child_base_smqf_uid = parent_base_survey.parent_base_survey_uid
INNER JOIN parent_base_ippa ON parent_base_survey.parent_base_survey_uid = parent_base_ippa.parent_base_ippa_uid
INNER JOIN parent_base_eac ON parent_base_ippa.parent_base_ippa_uid = parent_base_eac.parent_base_eac_uid
INNER JOIN parent_base_scas ON parent_base_eac.parent_base_eac_uid = parent_base_scas.parent_base_scas_uid
INNER JOIN parent_base_smqf ON parent_base_scas.parent_base_scas_uid = parent_base_smqf.parent_base_smqf_uid
INNER JOIN parent_base_eval ON parent_base_smqf.parent_base_smqf_uid = parent_base_eval.parent_base_eval_uid
INNER JOIN user_access_level ON parent_base_eval.parent_base_eval_uid = user_access_level.user_access_level_uid
INNER JOIN parent_one_month_survey ON user_access_level.user_access_level_uid = parent_one_month_survey.parent_one_month_survey_uid
INNER JOIN parent_one_month_ippa ON parent_one_month_survey.parent_one_month_survey_uid = parent_one_month_ippa.parent_one_month_ippa_uid
INNER JOIN parent_one_month_eac ON parent_one_month_ippa.parent_one_month_ippa_uid = parent_one_month_eac.parent_one_month_eac_uid
INNER JOIN parent_one_month_eval ON parent_one_month_eac.parent_one_month_eac_uid = parent_one_month_eval.parent_one_month_eval_uid
INNER JOIN child_three_month_survey ON parent_one_month_eval.parent_one_month_eval_uid = child_three_month_survey.child_three_month_survey_uid
INNER JOIN child_three_month_scas ON child_three_month_survey.child_three_month_survey_uid = child_three_month_scas.child_three_month_scas_uid
INNER JOIN child_three_month_smqf ON child_three_month_scas.child_three_month_scas_uid = child_three_month_smqf.child_three_month_smqf_uid
INNER JOIN parent_three_month_survey ON child_three_month_smqf.child_three_month_smqf_uid = parent_three_month_survey.parent_three_month_survey_uid
INNER JOIN parent_three_month_scas ON parent_three_month_survey.parent_three_month_survey_uid = parent_three_month_scas.parent_three_month_scas_uid
INNER JOIN parent_three_month_smqf ON parent_three_month_scas.parent_three_month_scas_uid = parent_three_month_smqf.parent_three_month_smqf_uid
INNER JOIN parent_three_month_eval ON parent_three_month_smqf.parent_three_month_smqf_uid = parent_three_month_eval.parent_three_month_eval_uid
INNER JOIN cron ON parent_three_month_eval.parent_three_month_eval_uid = cron.cron_uid WHERE registration.rego_parent_uid = ?");
$members->bindParam("s", "" . ((isset($_SESSION["rego_parent_uid"])) ? $_SESSION["rego_parent_uid"] : "") . "", "-1"); //WAQB_Param1
$members->execute();
?>

Answer 1

Run them as separate queries, from a cursory glance, it looks like you are just joining every possible table with data associated with a reference to an id value. Your results are going to be the cross product of every matching row in each table. For N tables, each row in table X will be repeated r0*r1*r2*....rX-1*rX+1*...rN. With the 23 tables there, if each only had 2 rows, you'd have nearly 8.4 million rows in your results.