web-programming | UiS DAT310 Web Programming course , spring

I am running a B2B website, where customers login, view the catalog and place orders. The website runs on a sub-domain (I think this is an important side note because of this). By the time customers finish their order and click on the place order button, they are sent back to the login screen, loosing their order. Sometimes, this period is shorter sometimes longer. Something to do with the Application Pool being restarted?.

So I created a logging functionality to see what's happening and came to find out whenever a user logs in, the following logs are created:

I am kind of new to SQL and web-programming right now, so I am doing a little project for myself to get to know every aspect of developing a website (yup, from frontend to backend and sql).

So the point is that potentially I may have a lot of data in my table, like over 3k rows [elements] (relatively a lot) with a bunch of columns [properties] as well. And I know from the beginning that it could be split on, say, four tables. For example, by color. Each element of each color has the same amount of keys and keys itself

So the question is how I could estimate the ratio of time:memory efficiency in this case. I do understand that it is much quicker to search for information in a smaller table, but I have no idea how do SQL tables are stored. For instance, how much additional memory each table costs without.