Dataset A - Join 2 tables
InvoiceHeader - InvoiceId, InvoiceDate
InvoiceDetail - InvoiceId, ProductId, Qty
These 2 tables are included in dataset and joined by InvoiceId.
Dataset B - Single table
Invoice - InvoiceId, InvoiceDate, ProductId, Qty
This single table is included in dataset.
What is the SPICE size of dataset A?
sum of (size of InvoiceHeader) and (size of InvoiceHeader) before the join
OR
size of the Join results (which is equivalent to dataset B) ?
SPICE stores the resulting data after the join, not the sum of the individual table sizes before the join.
Since Dataset A joins InvoiceHeader and InvoiceDetail using InvoiceId, the resulting dataset structure will be similar to Dataset B, containing one row per InvoiceId and ProductId.
This means Dataset A and Dataset B will have approximately the same SPICE size, assuming:
No additional transformations or filters are applied.
The number of records in the resulting join matches Dataset B.