I have a data set that needs to be regularly searched and having a performance issue with FMPro8.5v2Advanced.
The data consist of 15 pairs of numbers and the way the matching script is currently working the first record (30 numbers) is copied to 30 different variables, then a find is run with the first pair of numbers and all records in the table that contain the first two numbers get their match score incremented by 1, next a show omitted is run and all records that do not have the first pair of numbers get the non-match score incremented by 1. This process is repeated for the other 14 pairs of numbers. Finally a find is done for all records that have a match score above a given threshold which the user can vary depending on their needs.
The problem is that with 4000 records in the database the matching process takes around 10 seconds per record or 11 hours.
Does anyone know a faster way to accomplish this?