Answer Posted / guest
Conceptual Data Model
During the Planning phase of the project, the conceptual
data model is created to capture the high-level data
requirements for the project. Since the model captures the
highlights of the client’s information needs, it is the
only model that effectively reflects the enterprise level.
Depending on the requirements, the enterprise-wide vision
may need to be emphasized to help guide the client in the
development of an overall data warehousing strategy. Detail
models that reflect the project’s scope will be created
during logical and physical data modeling. The conceptual
data model is the precursor to the logical data model; it
is not tied to any particular solution or technology.
Entities, relationships, major attributes, and metadata
across functional areas are included. During successive
releases, the conceptual data model should be validated and
updated if necessary. An enterprise should have only one
conceptual data model.
Logical Data Model
During the design phase of the project, the logical data
model is created for the scope of the complete project. A
portion of the conceptual data model will be fully
attributed and completed as the logical data model. The
logical data model reflects the technology to be used. In
today’s environment, this typically means either a
relational DBMS or a multidimensional tool. But if the
client should be using an older DBMS such as IMS or IDMS,
the logical model will be quite different than if an RDBMS
is to be used. The logical data model reflects a logical
data design that can be used by the developers on the
project. For an RDBMS, that means logical tables (views)
and columns.
Physical Data Model
Like the logical data model, the physical data model is
created during the design phase. This modeling activity
should reflect the scope of the specific release of the
project. The model’s final design will be highly dependent
on the technical solution for the data warehouse. The
purpose of this model is to capture all the technical
details required to produce the final tables, and physical
constructs such as indexes and table partitions. The
logical data model will serve as a blueprint to the project
team while the physical data model is a blueprint for the
DBAs. All the functionality reflected in the logical data
model should be preserved while creating the physical data
model. The generated table schemas will be identical to the
physical data model.
| Is This Answer Correct ? | 11 Yes | 5 No |
Post New Answer View All Answers
In general, how many Fact & Dimension tables are present in a Banking Data warehouse project ?
What are the different types of data warehouses?
What are templates in which way it is help full?
What are the data types present in bo? What happens if we implement view in the designer n report?
Explain what are fact, dimension, and measure?
What are partition points?
What are data modeling and data mining?
What are situations where snow flake schema is better than star schema to use and when the opposite is true?
What is bi?
Explain what is the advantages data mining over traditional approaches?
Explain where do we use semi and non additive facts?
Explain what type of indexing mechanism do we need to use for a typical datawarehouse?
What are snapshots? What are materialized views & where do we use them?
What is Sequence clustering algorithm?
How to generate the cubes in framework manager?
