This paragraph lists some limitations and constraints but also has tips and guidelines from best practices and real-life cases.
An InfoCube’s dimensions can range from one to 32. Generally, the number of dimensions of an InfoCube should not exceed 7 or 8. An InfoCube with more than 8 dimensions can be difficult for end-users to understand and use. Before creating an InfoCube with more than 8 dimensions, consider revising your data model to reduce the number of dimensions.
An InfoCube can have up to 256 versions. Generally most InfoCubes only have a single version. Add other versions only when needed. When secondary versions are needed, generally four or five versions are sufficient. Only very rarely are more than 12 versions needed. If you have more than 12 versions, analyze whether or not each version is really needed.
The maximum theoretical number of cells of an InfoCube version that does not include a sparse structure is 10^28 cells (i.e. this is the virtual size limit of a non-sparse InfoCube version). To calculate the theoretical number of cells of an InfoCube version, multiply the MaxItemNumber of its entities. For time entities, the MaxItemNumber is equal to the number of members.
For an InfoCube version that includes a sparse structure: the product of the MaxItemNumber of non-sparse entities must be less than 7.8x10^28; the product of the MaxItemNumber of sparse entities must be less than 7.8x10^28 and the overall product of the MaxItemNumbers must be less than 10^37 (note that the given numbers are approximations).