Question

I want to add only unique records into a database table. Is there any way to do this without having to do a select to see if the current record already exists, because that will get very time consuming as the table grows. I am open to all types of suggestions (if they exist).

Also as a possible alternative, perhaps there are some indexing options that make these selects faster, or perhaps the data can be ordered in such a way as to make the select statement execute faster.

I am using MySQL and Java.

Answer 1

There are two possibilities.

1. Assume it’s not a duplicate, so perform an INSERT and cope with the err by doing an UPDATE.
2. Assume it is a duplicate, so perform an UPDATE and cope with the error by doing an INSERT.

Which is better depends on the relative probabilities.