I have added a time series measure in RPD
Year Ago Sales = AGO("Enterprise Analytics"."Facts Sales Fact"."Sales Le $" , "Enterprise Analytics"."Hierarchy Time"."Fiscal Year" , 1)
If I pull 'Year Ago Sales' in Answers it takes forever to run it - ( 5-6 minute) - The result is correct.
If I simply pull - "Enterprise Analytics"."Facts Sales Fact"."Sales Le $" - runs in 5 seconds.
Is thier any way I can optimize the timeseries functions?
Any response would be greatly appreciated!!
We are on Oracle Business Intelligence 184.108.40.206.5 version.
Edited by: ann 2012 on Mar 14, 2013 2:13 PM
This is why I often make an alias to the fact table, and join it to the time dimension using year - 1. To me the SQL comes out legible and usually performs better than the version that comes out using AGO.
1. Create an alias fact table (Year Ago) to pull last year value.
2. Extend your fact table to store another measure (last year sales)
3. Based on volume of granular data and query pattern on year ago measures, you may create aggregate fact tables.
hope this helps.