Answer Posted / guest
To design any OLAP model, first we need to understand what
OLAP system we are designing.
We have 2 models for Understanding and implementing project.
E-R Model - which is logical helps us understand components.
Dimension Model - which is practical helps us design OLAP
system.
E-R Model helps us to analyse various Entities related to
the project we are about to design and relation between
them. Its more of a rough sketch using pen and paper or
mind map.
Let me give you an example by taking a College system for
an instance
Entities to be considered for designing college may be -
Administration (Principals, HODS)
Transportation (of course College provides Transportation)
Teaching faculty
Non-Teaching (workers, peons)
Canteen and Stationary
Hostels ( in case of Residential Colleges)
Accounts
.... goes on..
We just understand various components related to the
college database construction and identify the relation
between college and them.
Once E-R model is done and we are clear with the components
involving in database design, we go with Dimensional model.
Analytic Workspace Manager is the primary tool for
creating, developing, and managing dimensional objects in
Oracle Database. Using this tool, we actually design the
data for the
This deals with creating Database with Facts and
Dimensions, Tables for each component.
- Design Dimensions for each component.
- Analyse what type of schema would suit your purpose
whether star/snowflake model.
- Analyse primary key/ foreign key relationships in these
dimensions.
- Creation of Dimensions, Cubes to load data into them.
- Once Cubes are Created , we go for multi dimensional
analysis on them to detail levels and hierarchies.
and save changes in the AWM.
There are various additional components to be designed.
Please find it in the below link
http://docs.oracle.com/cd/B28359_01/olap.111/b28124/cubes.ht
m
Is This Answer Correct ? | 0 Yes | 0 No |
Post New Answer View All Answers
What is a core dimension?
Explain when do we analyze the tables? How do we do it?
Could any one explain me how to create cascading prompts?
How much data hold in one universe.
What are the various tools? Explain name a few?
What is the limit to the number of sources and targets you can have in a mapping?
What is metadata in context of a datawarehouse?
Explain the difference between keylokup and joiner?
Explain can report net connect multiple datasource at a time in report creation time?
What is a relationship?
What are data validation strategies for data mart validation after loading process?
Suppose we want to display zero when null values coming to report how can we do that?
What are the different types of datawarehosuing?
If two relational tables are connected to sq trans, what are the possible errors it will be thrown?
Explain what type of indexing mechanism do we need to use for a typical datawarehouse?