Comments by kevin.goodman (Top 1 by date)

Kschuler, althought I think you are technically right unfortunately the request times out after several hours! I knew I was close but thanks for your help.

As I said the R table has over 35m rows and P table about 30k.

Partially using your solution and introducing a third table I am able to chunk the data but this isn't really a solution to my bigger problem.
If anyone out there is able to help with the performance of the request it would be greatly appreciated.
This is my solution to chunking the requests where the table fileSubmissions JOINS table R ON the REF column

SELECT r.FREF FROM R asr
JOIN filesubmission as fs ON fs.REF=r.REF
JOIN P as p ON p.ID=r.ID
where fs.id > 1000 AND fs.id < 2000
group by r.REF
HAVING count(DISTINCT p.TYPE) > 1;