When I run an append query to update employees' skills based on attendance in a class in which certain skills were taught, multiple records match the initial query, but the records that need to be updated are relatively few. So, for 1,500 records matching, perhaps only 20 employees, the ones who attended the class, will be updated. The rest of the rows of records of the other unmatched rows are "deleted," for lack of a better understanding or description, so that there are multiple large gaps in the auto numbered primary keys of the training records. As I add training records, these gaps grow quite a bit wider, from the 100s to now, the 1,000s, and soon, the 10,000s.
I'm not sure if I'm describing very well, but it's the difference in the record amounts indicated by the dialog boxes while appending information.
Should those "missing records," or gaps, be a cause of concern as long as the data I want to be saved is being saved? Which, it is being saved. It just seems that the autonumbered primary key is going to become huge, quickly. Is that a performance hindering circumstance?
Perhaps there would be a better way to write my append query. One thought was to exclude records that didn't have a training date within the last X amount of days, but that would mess things up in case someone needed to update records for training prior to that time. I've attached a design and sql copy of my append query.