Hello,
Right now I have a table with about 25,000 records in it. I need to script a looped Perform Find based on a text field that always begins with "JT" and is always followed by a 5-digit number (e.g. "JT23445" and "JT31662").
I expect this table to expand by about 20,000 records per year, which means it won't be too long before I'm up to 100,000 records in this beast. Will I be better off from a performance point of view if I create a new number field that shitcans the "JT" prefix, then use that new number field as the basis for my looped finds?
Of course I'd love to have this specific question answered, but if people want to use it as a springboard for a more general discussion of the nuts and bolts of how FileMaker indexing works, I'd welcome that.
Thanks in advance.
Dean
If you are doing s scripted process such as..
find records.
do stuff to them
repeat until all necessary records are processed.
Your script will take longer to finish as the record count goes up.
Time to execute both the find part and the do stuff to them part will increase.
100k records might not be that big of a deal unless the table is very "wide" ( lots of fields )
There may be a better way based on what the "do stuff to them" actually does.
if you have FM server you might be able to ship the process off to the server as PSOS or as a server scheduled script.
In general FM's find engine is pretty fast but I do seem to remember discussions herein about using Get ( UUID ) text as a primary key value v.s. a similar function that returns only numbers that suggested find on numeric is faster.