SAP BO Universe Performance tunning guidelines

-
Below are Universe performance tuning guidelines :
These guidelines need to followed with Webi performance tuning guidelines
given separately on this blog to get good performance reports.

• Always use aggregate functions i.e.  Sum, max, min etc with measure objects.

 • Set the proper query limits at the universe, especially “Limit size of result set to” & “Limit execution time to”. Setting the right parameters will result in a good universe by performance wise

 • Create aggregated tables for reports taking too much time to execute due to huge data set in underline tables.

 • List of values:
  • Create LOV’s on small lookup/master tables instead of actual fact table.
  • Avoid LOV on measure and date objects.
  • Remove LOV from non required objects i.e. keep LOV on objects which are going to be used in filters.
• Use conditional objects/filters on Universe level to filter out non required data from universe level instead of reports

 • Push complex calculation of objects to ETL level i.e. store the already calculated value in underline tables

 • Minimize the use of variables/formulas on report level. Push them to universe or database level.

 • Create database index on columns used in joining tables in universe.

 • Create database index on columns used as filters or going to be used as filters on report.

 • Use Index awareness for better performance in case of aggregate tables of multiple granularities.

 • Build lean queries for your documents
  • Design Universe tables and relationships such a that only required tables are used in generated query for specific requirement.
• Array Fetch Size Optimization
  • Array Fetch Size sets the maximum number of rows with each fetch from the database
  • AFS can decide the performance of the data returned by reports.

• Use Query stripping to remove unused objects from query.

 • Avoid use of Derived tables.

Comments

Post a Comment

Popular posts from this blog

Oracle to SAP HANA Date Conversions

-
Are you migrating your reports and dashboards from Oracle to SAP HANA?  Below are some ready to use date conversion SQL formulas which would save your significant amount of time. Please note that I have considered Financial year from 1-April to 31 March for year calculations. Also, number of days Calculations are considered till previous day. 1)   Current Financial year start e.g. 1-Apr-2017  ADD_MONTHS(ADD_DAYS(CURRENT_DATE, DAYOFYEAR(CURRENT_DATE) * -1 + 1), 3) 2)  Previous Financial Year start  ADD_YEARS(ADD_MONTHS(ADD_DAYS(CURRENT_DATE, DAYOFYEAR(CURRENT_DATE) * -1 + 1), 3), -1)  3) First date of current Month   ADD_DAYS(LAST_DAY(ADD_MONTHS(CURRENT_DATE, -1)), 1)   4)First date of current Month Previous year   Add_months(ADD_DAYS(LAST_DAY(ADD_MONTHS(CURRENT_DATE, -1)), 1), -12) 5)Number of days till date This Financial year  DAYS_BETWEEN(ADD_MONTHS(ADD_DAYS(CURRENT_DATE, DAYOFYEA...
Read more

SAP BO Universe Best Practice check list

-
Below are best practice guidelines for Universe development • Universe Name & Universe Description should be defined with proper Business Name and description in the Universe properties. • Class name should be started in INIT cap. • Objects name should be started in INIT cap. • Object should be placed in alphabetical order in the class. • Measure Objects in any class should be placed after the dimensions/detail objects. • Always define every object with meaningful description so that it helps user easily understand the purpose of it which avoids going through heavy documentation. • Measure objects should be created with aggregate functions like sum, max, min etc. This will ensure that only required aggregated data is returned to reports instead of whole set of raw data. • Create the context with proper business logic names. It will help you to find out what context is used in a particular report. • Always place any derived tables or isolated tables down ...
Read more