April 14, 201114 yr I'm importing some data from a legacy database, and in the process I'm doing some field cleanup, for example changing an ID field to strip leading zeros: ReplaceFieldContents ( table::ID ; GetAsNumber(Table::ID) ) I'm also cleaning up some non-key fields. Using FMA 11v3, on a database on my local SSD. What I'm finding is that replacing data in a non-key field (one that is not used in any relationships) is about 100x faster than replacing data in a key field. The table, and the TOG, are used in many relationships, but always on the many side of the relationship. I'm confused by this behavior, because I have only a single window open. I'm also pretty sure that I don't have any parent/child record deletion features enabled. So -- why would filemaker care whether the data I'm updating is in a relationship field or not? I would hope that when updating data in a child record, unless the parent record was being accessed in some way, there's no need for filemaker to follow the relationship graph... Is the slowdown simply because the field is indexed? If so, it seems like indexing is slower in FM11, or something has changed. Any ideas about how to speed this up? I could disable indexing before the imports? Or would doing the imports & updates on a TOG that is not used in any relationships help performance? [edited for clarity]
Create an account or sign in to comment