You need to recommend a cube architecture for CUBE1. The solution must meet the performance requirements for CUBE1. Which two partitions should you recommend creating? Each Answer presents part of…

Partitions filtered by customerID values from the dimension table allow the storage engine to skip entire partitions when queries filter on the Customer dimension, directly matching the cube's dimension hierarchy traversal and reducing I/O.

Filtering partitions by customerID directly from the fact table's foreign key column does not align with the dimension hierarchy navigation the query engine uses, so partition elimination is less effective than when dimension table attribute values define the boundary.

Filtering partitions by productID from the fact table's foreign key column has the same limitation - partition boundaries defined on raw fact-table keys are not directly leveraged during dimension-hierarchy-based query pruning in SSAS.

Partitions filtered by productID values from the dimension table similarly allow partition elimination for product-based queries, and referencing the dimension table ensures partition boundaries are consistent with dimension member ranges used in MDX queries.