ASO Input and Aggregate Data Size

Hello,

We are implementing an EPCM solution and are concerned about data generation and storage with ASO cubes. Primarily, we would like to better understand what contributes to data size and if the growth is linear based purely on number of input cells. Additionally, does aggregate data size have the same relationship?

For example, for a single month of data, our cube contains 1.3 billion input cells and 31GB of input data size. Subsequent months will contain a similar level of detail. In theory, would adding a second month result in 2.6 billion input cells and would that correlate to 62GB of input data size? In this example, optimizing aggregation produces 3.3 billion aggregate cells across 16 aggregate views, which results in 79GB of aggregate data size. Would adding a second month result in 6.6 billion aggregate cells and 158GB of aggregate data, or are there efficiencies gained at the aggregate level?