Perfomance for Dropping a Column

I am working on some experimental routines to analyze hash functions. The big idea is to generate a lot of hashes. Of course I am storing them in a database table. For a while, I have been going back and forth trying to determine how to identify each hash.

Initially I think I just has a key that was generated from a sequence. Then I moved to multiple keys that were related to the input data. However I am changing how I generate that input data. So I thought I would drop those columns. Oops. It is taking a very long time. I only have 20M rows in the table. What is going on here?

Now I realize that I am going to regenerate all my hashes. So those existing 20M rows are just going to be truncated anyway. I guess I should have truncated first, then dropped the columns. You live, you learn. Still want to figure out why it is taking so long to drop a column here. Yeah there are a bunch of rows, and some of the columns may be big. But come on.