Hi Everyone!

I'd love to yap and chat, however I am now a busy businesswoman and have things to do, so I am making a page to review the material for my certification exam. Have you ever heard of an OLAP cube. Well now you fucking have. Im gonna tell you EVERYTHING you need to know (or will absolutely never need to know) about multidimensional databases!!

So a multidimensional database is a structure to hold information. But it has more than 2 dimensions, its not just a flat grid. Theres data under the data. We think of this as the data being in a cube.

Look at this sexy thing:

FOR EXAMPLE:

When we need data we just slice right into that delicious thing and get a view of its insides. We can configure views to access specific data points. the cube analogy is kinda stupid bc you can have like 13 dimensions but whatever. In theory its a hypercube but I dont make the rules.

Dimensions are just categories of data, but they have a hierarchy to them. Like if you have year data, some dude might be like wait I wanna see it more specific lets drill down to month and day and hour. And as the data guy you're like fuck you but also okay I can do that. This makes it easy to aggregate data down the line.

Okay so now we get the whole data structure thing. But most people dont. So thats why we put it into a dashboard or infolet. On a dashboard the user can select the dimensions they wanna see data on and it just magically updates. Wow! Infolets are just little widgets that show a tiny snapshot of data.

So what the fuck are people even using this data for. Businesses like to make money and whatnot so oftentimes they will do what we call scenario planning. We would likely have actuals which is just the literal past business data, which is used to make the budget, or planned numbers, which then drives the forecast, the projected numbers. We can make a "sandbox" in oracle where we can "play" with numbers to make scenarios and not fuck everything up if we make a terrible forecast. Sandboxes are completely isolated from live or actual data.