so what is a SAS OLAP n-way anyway?

2

In SAS OLAP the N-WAY summarization is created by default, however you have the option to not create the N-WAY & as I highlighted in a prior post, not creating it changes your OLAP type to either a HOLAP or ROLAP. But the real question is what the heck is this N-WAY thingy anyhow? For new SAS programmers this is probably a new term, but for SAS PROC SUMMARY and MEANS users this is ol' news. (In the SAS OLAP Documentation, it is actually stated that these two procedures were where the N-WAY term came from.)

The N-WAY option simply provides the lowest level summary of data. So by default, OLAP creates an N-WAY to house the lowest level of measures at ALL dimensional/hierarchy crossjoins.

*for example*

If you have 3 dimensions - geography, product, business unit
with several levels within  - state/city, color/size/name, organization/department
and measures of              -  sales sum, returns sum, inventory sum

The N-WAY is the total sum of sales, returns and inventory at the cross join of all levels:
state-city-color-size-name-organization-department.
Such as:

NC Raleigh Blue L Shirt Acme Mens $1.25 $2 $10
NC Raleigh Blue L Shirt Acme Womens $3.42 $4 $20

...

In a report, if you'd like to see the total sales by city and organization, SAS takes the nway and on-the-fly sums up the 'lower' level data at the departments (as well summing up the product dimension) to give you the totals.
Such as:

NC Raleigh Acme $4.67

Understanding this is important, because you can quickly realize the need for custom aggregations that automatically sum up the data to different crossjoins of dimensional levels. I recommend at least running ARM logging of usage in initial report testing to pull back into SAS OLAP's Advanced Aggregation Tuning plug in and automatically create some custom aggregations to improve performance.
Share

About Author

Angela Hall

Senior Technical Architect

Angela offers tips on using the SAS Business Intelligence solutions. She manages a team of SAS Fraud Framework implementers within the SAS Solutions On-Demand organization. Angela also has co-written two books, 'Building BI using SAS, Content Development Examples' & 'The 50 Keys to Learning SAS Stored Processes'.

2 Comments

  1. Pingback: Tuning OLAP Cubes - Real BI for Real Users

Back to Top