Code: CS482                      Subject: DATA WAREHOUSE DESIGN & IMPLEMENTATION

Flowchart: Alternate Process: AUGUST 2009Time: 3 Hours                                                                                                     Max. Marks: 100



·      Question 1 is compulsory and carries 28 marks. Answer any FOUR questions from the rest.  Marks are indicated against each question.

·      Parts of a question should be answered at the same place.



Q.1      a.  Differentiate between the operational data & Decision Support System data.


             b.  Explain how the system development life cycle for the data warehouse is exactly opposite to the classical SDLC.


             c.  Explain the four levels of data in the architectural environment.


             d.  Is the data in data warehouse homogenous or heterogeneous? Illustrate with an example.


             e.  What is granularity? Why it is important in data warehouse?


             f.   Explain ‘Normalization’ in data warehouse. List its advantages.                                          


             g. List the changes to be made in corporate data model when it is to be applied to the data warehouse.                                                                                                                             (7  4)


Q.2       a.  Explain the term ‘Living Sample Database’ in detail.                                                          


             b.  What is “Partitioning of data”? Illustrate with an example. Explain the ways to carry it out.   Which one is better and why?                                                                                                                (8+10)


  Q.3     a.  Explain how process model and data model can apply to the architecture environment? Why Process model is not suitable for data warehouse.                                                                                        


             b.  Explain in detail all the three data warehouse data models.                                       (8+10)


  Q.4     a.  Discuss star join with an example. Creating a star join for the data warehouse is               a mistake. Comment.                                                                                                                                        


             b.  Explain Multidimensional DBMS. Discuss the ways of implementing it  by providing its   strength and weaknesses. How is Multidimensional DBMS different from warehouse?                 (8+10)



  Q.5     a.  Building the warehouse on multiple levels is easiest scenario to manage, with fewest risks. Explain.  


             b.  Discuss in detail “Design review”.                                                                            (8+10)


  Q.6     a.  Discuss the technique of Event Mapping.                                                                           


             b.  Explain how the data warehouse acts as a basis for EIS.                                          (8+10)


  Q.7          Write short notes on any THREE:                                                                                    


(i)                                                                                                                                                                                                                                                         Cyclicity of Data

(ii)                                                                                                                                                                                                                                                       Drill Down Analysis

(iii)                                                                                                                                                                                                                                                             Importance of  metadata in data warehouse

(iv)                                                                                                                                                                                                                                                            Local data  V/s  Global data                                                                                           (6+6+6)