SSAS optimsiation

*	Tables	Many to many - Check tables are written optimally (no duplicates and < 1mill rows)
1	DSV	Remove any unnecessary tables
2	Roles	Phoenix role doesn't have permissions?
3	Dimension Usage	Manufacturer to Product	*Make sure all materialized
4	Dimension	Account - Attribute relationship	*Rigid, etc.
5	Dimension	Misc. Account - Attribute relationship
6	Dimension	Product - Attribute relationship
7	Dimension	Wholesaler - User Hierarchy?
8	Dimension	Attribute hierarchy enabled for all attributes	E.g. Not needed if in User Hierarchy
9	Dimension	Estimate row count for all attributes
10	Calculated script	Review all iff null and use is empty
11	Calculated script	Review all = make sure use object reference
12	Partitions	Create all partitions (5-20million rows)
13	Partitions	Specify slices for each partition
14	Aggregation	Initially set most common attributes used in group by
15	Aggregation	Switch on usage based optimisation and run all queries for reports - rebuild aggregations
16	SSIS Cache	Build cache warmer SSAS job to run most common top level queries after cube has been refreshed
17	Reports	Shell use security to get wholesaler role and territory - sub reports remove all parameters definitions, pass through from parent report and then use separate defined data source. Finally cache all sub reports run so will be instant	Can set up schedule for cache and clearing down