1. 8 Copyright © 2006, Oracle. All rights reserved. Previewing Advanced Oracle OLAP Features

2. 8-2 Copyright © 2006, Oracle. All rights reserved. Objectives After completing this lesson, you should be able to do the following: Discuss how the OLAP platform enables you to develop BI solutions that extend beyond the analytic capabilities exposed through AWM Explain the roles of the different AWM views Access OLAP Worksheet Describe some of the advanced calculation capabilities of Oracle OLAP Identify the purpose of Oracle OLAP models Enable your AW for very simple SQL access, and understand how this feature can be exploited

3. 8-3 Copyright © 2006, Oracle. All rights reserved.

4. 8-4 Copyright © 2006, Oracle. All rights reserved. Oracle OLAP Platform: Features for IT Statistical Functions Financial Functions Custom Functions Conditional Logic Looping OLAP DML: Inspects the physical implementation of the AW Leverages OLAP’s prebuilt advanced analytic functions Defines procedures or functions that require programming, including conditional logic and looping OLAP AW API: Is a Java API that enables you to define a complete AW and manage its life cycle Is used to develop the AW model, and not just for the physical implementation Exposes key OLAP functionality Analytic Workspace API Multidimensional Oracle Database 10g Analytic Workspace Manager 10g (Model View) Warehouse Builder 10g Models Time-Series Functions Forecasts Aggregation Functions Allocations Custom Solution

5. 8-5 Copyright © 2006, Oracle. All rights reserved.

6. 8-6 Copyright © 2006, Oracle. All rights reserved. AWM 10g Viewing Options Model view: –AWM 10g applies Standard Form. –Standard Form is a requirement for Oracle Business Intelligence tools and applications. Object view: –Not all AW have (or need) Standard Form. –You can view all physical objects in an AW by using the Object view.

7. 8-7 Copyright © 2006, Oracle. All rights reserved.

8. 8-8 Copyright © 2006, Oracle. All rights reserved. Using the AWM 10g Object View The Object view provides: A sorted list of all objects in the AW Ability to add, delete, or change objects * * Use with caution

9. 8-9 Copyright © 2006, Oracle. All rights reserved.

10. 8-10 Copyright © 2006, Oracle. All rights reserved. Using the OLAP Worksheet Interface Use OLAP Worksheet to interface with OLAP DML and SQL commands and programs. * To start OLAP Worksheet, select Tools > OLAP Worksheet. * Use with caution

11. 8-11 Copyright © 2006, Oracle. All rights reserved. Entering OLAP DML Commands Query/Command window Response window Command History/Log [F2] OLAP DML Help [F1] Clear Response window Clear Command window Execute Commands* *This button is unavailable if the “Execute on Enter” box is selected.

12. 8-12 Copyright © 2006, Oracle. All rights reserved. Using SQL in the OLAP Worksheet To enter SQL commands in the OLAP Worksheet, select the SQL Mode option. 1 2 3

13. 8-13 Copyright © 2006, Oracle. All rights reserved. Using the OLAP Worksheet Help System 1 2

14. 8-14 Copyright © 2006, Oracle. All rights reserved. Using OLAP Help: Index Tabbed Page 1 2 3

15. 8-15 Copyright © 2006, Oracle. All rights reserved. Oracle OLAP Calculation Capabilities Oracle OLAP provides a wide range of calculation types: Calculations along dimensions: –Aggregation –Allocation –Models Calculated measures: –Calculated measures by using the Calculation Wizard –OLAP expressions used in calculated measures

16. 8-16 Copyright © 2006, Oracle. All rights reserved.

17. 8-17 Copyright © 2006, Oracle. All rights reserved. Oracle OLAP Models Models are calculation rules that operate across the members of a dimension. Models are commonly used in financially oriented OLAP systems, but are useful in many other cases too. In models, some dimension members have their measure values entered, whereas others are calculated. Unlike aggregation and allocation, models allow for different dimension members to be calculated independently of any hierarchy, using a wide range of calculation logic.

18. 8-18 Copyright © 2006, Oracle. All rights reserved. Models: An Example on a LINE Dimension 1 2 3 4

19. 8-19 Copyright © 2006, Oracle. All rights reserved.

20. 8-20 Copyright © 2006, Oracle. All rights reserved. Using Models 1 23

21. 8-21 Copyright © 2006, Oracle. All rights reserved. Benefits of Models Models are stored in the AW so that all authorized users of the AW can access them using any tool. A model evaluates all the equations in the correct order, including the solving of simultaneous equations. A model can be applied to any measure that has the modeled dimension (such as LINE ) as one of its dimensions. Models can be executed dynamically at query time, in advance, or on command by an application.

22. 8-22 Copyright © 2006, Oracle. All rights reserved. Understanding More About Calculated Measures

23. 8-23 Copyright © 2006, Oracle. All rights reserved. Calculated Measures Using Custom Expressions Example: Conditional Logic Based on an Attribute if LINE.VARIANCETYPE.ATTRIBUTE eq ‘POSITIVE’ then FINANCIAL_DATA.ACTUALS.MEASURE – FINANCIAL_DATA.PLAN.MEASURE else FINANCIAL_DATA.PLAN.MEASURE – FINANCIAL_DATA.ACTUALS.MEASURE 1 2 3

24. 8-24 Copyright © 2006, Oracle. All rights reserved.

25. 8-25 Copyright © 2006, Oracle. All rights reserved. Using Custom Expressions to Alter Display Behaviour Earlier, you created measures in the (3D) Budgets cube and the (4D) Sales cube. You have seen how different shaped measures can be displayed together. But what if a user drills up or down on Total Customer? Budget does not vary by Customer and so is assumed to be the same for all customers. It makes no business sense to drill up or down on budget, but you can do that. Misleading?

26. 8-26 Copyright © 2006, Oracle. All rights reserved. Using Custom Expressions to Alter Display Behaviour Example Solution Create a 4D “formula” over the 3D measure. Use custom expressions to qualify how and whether data should be displayed for the Customer dimension. Calculation Expression: if CUSTOMER.DIMENSION eq 'TOTAL_1' then BUDGETS.BUD_SALES.MEASURE else NA Problem Solved Use this controlled 4D version of Budget measures in all other calculations.

27. 8-27 Copyright © 2006, Oracle. All rights reserved. Using OLAP Calculations in Practice In practice, you find that: You need to define many calculations You often want to define similar types of calculations across base measures and calculated measures Calculation types are often similar across solution areas For each of: Sales, Cost, Profit, Quantity, Budget Sales, Budget Cost, Budget Profit, Profit Margin, Average Sales Price, etc You require many similar calculations : Last Yr, LY %, LY Dif; Prior Pd, PP %, PP Dif; YTD etc Share of Total, Share of Parent (for each dimension) etc Comparison to Budget, to Target, to Plan etc etc

28. 8-28 Copyright © 2006, Oracle. All rights reserved. Using Excel to Define OLAP Calculations Available for download from OTN A sample using AW API to implement measures in bulk Each row representing an OLAP calculation definition Easy to define many calculations quickly (Cut/Paste) Templates and free-form calculations supported

29. 8-29 Copyright © 2006, Oracle. All rights reserved. The Spreadsheet Calculations Sample (download available on OTN) Demonstration Use E:\Labs\Viewlets\Using_OLAP_viewlet_swf.html for a viewlet demonstration (approximately 7 minutes). Use E:\Labs\SpreadsheetCalcs\ OLAPCalculations_UsingOLAP.xls for the spreadsheet used to enhance SALES_AW in SH_AW_TUTOR.

30. 8-30 Copyright © 2006, Oracle. All rights reserved. Excel Leverages the AW API Analytic Workspace API RelationalMultidimensional Oracle Database 10g Oracle OLAP 10g Administration Analytic Workspace Manager 10g Warehouse Builder 10g Custom Solution Excel with OLAP Calculations

31. 8-31 Copyright © 2006, Oracle. All rights reserved. Excel Leverages SQL Access to Oracle OLAP Standard Form AW Metadata The Excel sample uses drop-down lists with valid inputs. Fields are populated by querying Oracle OLAP Active Views. OLAP Active Views describe the metadata found in an analytic workspace. Any tool that can query Oracle with SQL can do this.

32. 8-32 Copyright © 2006, Oracle. All rights reserved. Metadata Retrieved via Views List of analytic workspaces: select owner || '.' || AW as AW from all_olap2_aws where SF_VERSION is not null List of cubes: select aw_logical_name from all_olap2_aw_cubes where aw_name='" & strAW & "' and aw_owner='" & strOwner & "'" List of measures: select aw_measure_name from all_olap2_aw_cube_measures where aw_name='" & strAW & "' and aw_owner='" & strOwner & "'" Using VBA and Microsoft ActiveX Data Objects, retrieve: … and more

33. 8-33 Copyright © 2006, Oracle. All rights reserved. User-Defined Functions User-defined functions can be created when the Oracle OLAP prebuilt functions do not meet your specific calculation requirements. You can use OLAP DML to: –Create the required function –Use the customized function in the expression of a calculation

34. 8-34 Copyright © 2006, Oracle. All rights reserved. Creating User-Defined Functions 1. Create an OLAP DML function:

35. 8-35 Copyright © 2006, Oracle. All rights reserved. Creating User-Defined Functions 2. Create a measure via the AW API whose equation is the function:

36. 8-36 Copyright © 2006, Oracle. All rights reserved. Creating User-Defined Functions 3. Use your new measure or function:

37. 8-37 Copyright © 2006, Oracle. All rights reserved. SQL Access to Analytic Workspaces By having SQL access to the AW, you can: Add value to SQL-based tools and applications Leverage existing SQL-based tools and applications Use multidimensional data with any other data—in a single tool, view, or query: relational, spatial, and so on Avoid duplicating data between multidimensional and relational databases Simplify application development: –Building AWs is easy. –SQL on AWs is easy.

38. 8-38 Copyright © 2006, Oracle. All rights reserved.

39. 8-39 Copyright © 2006, Oracle. All rights reserved. Simple SQL Access to Analytic Workspaces SELECT time_desc, channel_desc, product_desc, geography_desc, sales, sales_ly_pct, sales_pp_pct, sales_shr_parentprod FROM sales_cubeview WHERE product_colour = 'RED', AND geography_level = 'REGION', AND time_level = 'MONTH', AND time_parent = '2002'; Simple SQL No joins No SQL calculations No SQL aggregations SELECT time_id, channel_id,product_id, geography_id, sales, costs FROM sales_cubeview WHERE product_id = '123', AND geography_id = 'ABC', AND time_id = 'JUL02';

40. 8-40 Copyright © 2006, Oracle. All rights reserved.

41. 8-41 Copyright © 2006, Oracle. All rights reserved. Degrees of Interaction with the AW Select from a view over the AW: –Access to AWs is transparent to the application Embed OLAP expressions in SQL: –Enables selecting from OLAP DML functions –Often supported by existing applications without explicit support for AWs Select from the OLAP_TABLE table function: –Views are not required. –Enables dynamic mapping between columns and AW objects –Can include OLAP DML commands

42. 8-42 Copyright © 2006, Oracle. All rights reserved.

43. 8-43 Copyright © 2006, Oracle. All rights reserved. SQL Access to AWs: Key Points The AW presents data to the SQL tool as if it is: –Fully calculated –Fully aggregated –Prejoined (even if the AW is actually doing it at query time) The SQL Access feature promotes very simple SQL. It provides access to any calculation: –Time series, shares, indexes, forecasting, modelling, and so on –Nonadditive aggregations and allocations Select the rows and columns needed.

44. 8-44 Copyright © 2006, Oracle. All rights reserved. Summary In this lesson, you should have learned how to: Explain how the OLAP platform enables you to develop BI solutions that extend beyond the analytic capabilities exposed through AWM Use the Object and Model views in AWM Access OLAP Worksheet Execute a command in OLAP Worksheet Describe how Oracle OLAP models further extend the calculation capabilities of the AW Explain how simple SQL may be used to access the power and performance of the analytic workspace

45. 8-45 Copyright © 2006, Oracle. All rights reserved. Oracle Database 10g: Using OLAP Course Summary In two days, you have learned: OLAP concepts Oracle Database 10g OLAP product features The logical dimensional data model for OLAP How to build Oracle OLAP multidimensional analytic workspaces How to tune AWs for large volumes of sparse data How to create calculated measures How to use OracleBI Spreadsheet Add-In How to use OracleBI Discoverer OLAP The availability of additional deep analytical power and functionality for OLAP

46. 8-46 Copyright © 2006, Oracle. All rights reserved.